Printer Friendly

Benefits and risks of using gelatin solution as a plasma expander for perioperative and critically ill patients: a meta-analysis.

The importance of intravenous fluids in the resuscitation of hypotensive patients was first recognised over a hundred years ago. Debate surrounding what type of fluid to use has continued for nearly as long and remains unresolved. Although colloid solutions may improve haemodynamic responses better than crystalloid solutions, a recent large randomised controlled trial (RCT) of using isotonic albumin in critically ill patients failed to demonstrate significant improvement in patient outcomes compared to isotonic saline solution (1). Furthermore, whether one colloid solution is superior to the others remains uncertain.

Gelatin used for intravenous infusion is usually extracted from bovine collagen. Its use in resuscitation was first described in 1915, but it was not until the Second World War that interest in using gelatin solution as a plasma expander was renewed and further trials showed that gelatin solution was effective in increasing plasma volume for patients in shock (2,3). The two common forms of gelatin solutions in clinical use today are urea-linked gelatin or polygeline (Haemaccel[R], Piramal Healthcare Ltd, Mumbai, Maharashtra, India) and succinylated gelatin (Gelofusine[R], B. Braun Melsungen AG, Melsungen, Germany). Both intravenous gelatin-based solutions have a plasma half-life of around three hours, and are relatively cheap and easy to store.

Anaphylactic and anaphylactoid reactions, with a quoted incidence of 0.038%, appear to be the most common serious complication of gelatin solutions (4), but concerns about their effects on coagulation have also been raised (5,6). We hypothesised that intravenous gelatin is safe as a plasma expander and conducted a meta-analysis to assess whether gelatin solutions would result in excess mortality, a higher risk of bleeding requiring transfusion or higher rates of acute renal failure compared to a) intravenous crystalloid, b) starch, c) isotonic albumin solution or d) dextran in perioperative and critical care settings.

METHODS

Literature search

Two authors independently searched the biomedical online database MEDLINE[R], National Library of Medicine, Bethesda, Maryland, USA, (1948 to December 2010) and EMBASE[R], Reed Elsevier, London, UK, (1980 to December 2010) for relevant RCTs comparing gelatin solutions with a) crystalloids, b) starch, c) isotonic albumin or d) dextran. The following Medical Subject Headings were used: "polygeline" or "gelatine" searched for either alone or in combination with "shock", "hemorrhage", "fluid therapy" or "sepsis". We also searched for the terms "haemaccel", "gelofusine" and "major surgery". Full text articles were obtained for any comparative studies we identified which used a gelatin-based solution as a plasma expander in humans. There were no language restrictions. Non-English articles were translated by speakers fluent in the language and with a medical background. Both paediatric and adult studies were included. Reference lists of included trials were also searched for relevant articles. Studies which were not randomised (e.g. cohort studies), allowed crossover, used blood products as the comparator fluid or were performed in vitro or in healthy volunteers, were excluded.

Data extraction

The primary outcome of this review was hospital mortality (or 28-day mortality when hospital mortality was not available). Other outcomes were incidence of acute renal failure (as defined by the original study), total blood loss during the study period, proportion of patients requiring allogeneic blood transfusion, volume of blood transfused (assuming one unit = 300 ml), incidence of allergic reactions and length of hospital stay. All outcomes and definitions were predetermined prior to the commencement of data extraction. Two authors independently extracted the data using a standardised data abstraction form and we clarified any missing or ambiguous data of the included studies by contacting relevant corresponding authors. In one paper, relevant data were extracted from graphs, as we could not ascertain the data from the tables or from the authors of the study (7). A number of studies were found to describe the same study population--in these cases we combined all the published manuscripts derived from the same cohort of patients as one study (8-11,29,30). Any discrepancies in data extracted were resolved by consensus amongst the three investigators.

Statistical analysis

Continuous data were described as mean and SD. If only median and range were reported, the median was used to approximate the mean, and range divided by four was used to estimate SD. Data entry was performed by one of the investigators and values subsequently checked for transcription errors by a second investigator. Analyses were performed using the software Review Manager[R] version 4.3.2 (Cochrane Collaboration.12, Nordic Cochrane Centre, Copenhagen, Denmark). The analyses were initially performed using a fixed-effect model and, if the results were significant, repeated with a random-effects model to confirm robustness of the results. Summary effects were expressed as odds ratios (OR) with 95% confidence intervals (CI) for dichotomous variables and weighted-mean-differences (WMD) with 95% CIs for continuous variables, stratified by the type of comparator fluid or type of starch used ('old' vs newer 6% hydroxyethyl starch 130/0.4).

The presence of heterogeneity was assessed by the [I.sup.2] statistic and an [I.sup.2] >40% was taken as significant heterogeneity in this study. A restricted analysis on the mortality effect of gelatin solutions in cardiac surgical patients alone or studies that had adequate allocation concealment and double-blinding was also performed. Using mortality or proportion of patients requiring allogeneic blood transfusion as an end-point, two funnel plots were used to assess the risk of publication bias. A P value of less than 0.05 was considered to be significant.

RESULTS

Study selection

Of the 66 studies on gelatin solutions identified from the literature search, 30 studies met the inclusion criteria (7-9,13-40) and were subject to meta-analysis (Figure 1). Thirteen studies recruited patients undergoing elective cardiac surgery, nine studies recruited patients undergoing orthopaedic or major abdominal surgery, one study recruited patients after trauma, one study recruited paediatric patients with septic shock and the remaining six studies recruited critically ill adult patients in the intensive care unit. A total of six studies had adequate allocation concealment, three studies were double-blinded, but none had both adequate allocation concealment and double-blinding. The characteristics of the included studies are described in Table 1.

[FIGURE 1 OMITTED]

Effect of using gelatin on risk of mortality

A total of 14 trials involving 1788 patients reported data on mortality. Using gelatin as a plasma expander was not associated with an increase in mortality when compared to other types of intravenous fluids when all strata of different types of alternative fluids were pooled together (OR 1.03, 95% CI 0.80 to 1.32) (Figure 2). The effect of gelatin on mortality was not different when compared to only a) crystalloids (OR 1.22, 95% CI 0.75 to 1.99), b) hydroxyethyl starches (OR 0.91, 95% CI 0.59 to 1.41), or c) albumin (OR 1.01, 95% CI 0.67 to 1.51). There was no significant heterogeneity between the pooled trials ([I.sup.2]=0%).

Effect of gelatin on risk of acute renal failure

A total of five trials involving 431 patients reported data on incidence of acute renal failure. Overall there was no significant difference between gelatin and alterative intravenous fluids on the risk of acute renal failure when all strata of different alternative fluids were pooled together (OR 0.74, 95% CI 0.42 to 1.29) (Figure 3), but significant heterogeneity between the included trials was observed ([I.sup.2]=55.4%). When compared to intravenous hydroxyethyl starches, using gelatin solutions was associated with a lower risk of acute renal failure (OR 0.43, 95% CI 0.20 to 0.92), primarily due to the data from one study using older starch in septic intensive care unit patients (27). There were no significant differences in the risk of acute renal failure when gelatin was compared to crystalloids (OR 0.73, 95% CI 0.14 to 3.89) or isotonic albumin (OR 2.20, 95%CI 0.71 to 6.81).

Effect of gelatin on transfusion requirement

A total of 11 studies comprising of 796 patients reported on the amount of allogeneic blood transfusion during the study period. There was no significant difference between gelatin solutions and other intravenous fluids when all strata of different alternative fluids were pooled together (WMD 0.0 ml, 95% CI -0.51 to 0.51; [I.sup.2]=28.6%) (Figure 4). There was, however, a significant difference in the amount of allogeneic blood transfusion in subgroup analysis when gelatin was compared to isotonic albumin solution alone (WMD 180 ml, 95% CI 8.1 to 353.6). Transfusion requirement was not significantly different after using gelatin as a plasma expander compared to crystalloids (-24 ml, 95% CI -91.2 to 42.2), starch (0 ml, 95% CI -0.51 to 0.51), or dextran (100 ml, 95% CI -326.8 to 526.8).

Effect of gelatin on risk of requiring allogeneic blood transfusion and blood loss

Although the proportion of patients requiring allogeneic transfusion after receiving gelatin solutions was not significantly different from patients who had received other types of intravenous fluids when all strata of different alternative fluids were pooled together (OR 1.04, 95% CI 0.77 to 1.40), more patients in the gelatin group required allogeneic transfusion compared to patients given crystalloids (OR 1.67, 95% CI 1.02 to 2.73), explaining the heterogeneity between the strata of trials (I2=42.3%) (Figure 5). When this subgroup analysis was repeated with a random-effects model this difference did not reach statistical significance (OR 1.58, 95% CI 0.95 to 2.61). There was no difference seen between gelatin vs starch (OR 0.79, 95% CI 0.53 to 1.17), or gelatin vs isotonic albumin (OR 0.5, 95% CI 0.1 to 2.63).

Nineteen trials involving 1252 patients reported total volume of blood loss during the study period. There was no significant difference in blood loss between those who were treated with gelatin solutions and other intravenous fluids, when all strata of different alternative fluids were pooled together (WMD 7.56 ml, 95% CI -18.75 to 33.87). This was consistent across different strata of comparisons with minimal heterogeneity (crystalloid: WMD 27 ml, 95% CI -14.7 to 68.7; albumin: WMD 89 ml, 95% CI -34.6 to 213.0; hydroxyethyl starches: WMD -12 ml, 95% CI -48.1 to 22.4, I2=18.8%) (Figure 6).

Other outcomes

Three studies reported hospital length of stay of 262 patients. No significant difference between gelatin solutions and other intravenous fluids was found (WMD 0.27 days, 95% CI -0.61 to 1.15; [I.sup.2]=0%). Five studies reported the incidence of allergic reactions and none of the 229 patients randomised to receive gelatin solution experienced any adverse reactions.

Sensitivity analyses

A reduced risk of acute renal failure after gelatin solutions compared to starches persisted, but the difference in proportion of patients requiring allogeneic transfusion became insignificant (OR 1.58, 95% CI 0.95 to 2.61) when the data were analysed by a random-effects model.

[FIGURE 7 OMITTED]

[FIGURE 8 OMITTED]

In the restricted analyses, the mortality of cardiac surgical patients was not significantly different between patients who were treated with gelatin solutions and other types of intravenous solutions. Because no studies had both adequate allocation concealment and double-blinding, a restricted analysis on higher quality studies was not possible.

Publication bias

Using mortality and the proportion of patients requiring allogeneic blood transfusion as end-points, the funnel plots did not suggest significant publication bias favouring gelatin solutions in small studies (Figures 7 and 8).

DISCUSSION

Although using colloid solution has a theoretical advantage of generating a greater intravascular expansion than crystalloid solutions (41), it also has the disadvantages of greater cost and the risk of adverse reactions and potential adverse effects on the coagulation and renal systems. Our results showed that using gelatin solution as a plasma expander was not associated with either a beneficial or an adverse effect on mortality compared to other types of intravenous fluid, including crystalloids, starch or isotonic albumin. Gelatin was however, associated with a lower risk of acute renal failure compared to older starch. When compared to isotonic albumin and crystalloids, gelatin appeared to be associated with a larger amount of transfusion and a higher risk of requiring allogeneic blood transfusion, respectively.

The optimal choice of intravenous fluid replacement therapy for perioperative and critically ill patients remains highly controversial, especially after many studies on colloid solutions published by Boldt were retracted recently. After excluding all the studies that were retracted, our results showed that using gelatin solution as a plasma expander was associated with a lower risk of acute renal failure compared to hydroxyethyl starches (OR 0.43, 95% CI 0.20 to 0.92) (Figure 3). However, this subgroup analysis consisted of data from only two studies with a total of 219 patients and cases of acute renal failure occurred predominantly in the study comparing 'older' starch with gelatin in critically ill patients with sepsis (27). No case of acute renal failure was observed in a small study (n=90) comparing gelatin to the 'newer' starch (25). As such, gelatin may still be preferable to starch, especially in patients who are at risk of developing acute renal failure, at least until adequately powered RCTs have confirmed the safety of newer starches.

The association between the use of gelatin and a higher risk of requiring allogeneic transfusion compared to crystalloids (OR 1.67, 95% CI 1.02 to 2.73) (Figure 5), or larger transfusion requirement compared to albumin (WMD 180 ml, 95% CI 8.1 to 353.6) (Figure 4) requires careful consideration. On the one hand, a number of other studies have demonstrated that gelatin solution can reduce maximum in vitro clot strength, as measured by thrombelastography and by fibrin polymerisati on (8,9,18,19,22,42). On the other hand, we did not find an association between gelatin and total blood loss compared to other intravenous fluids (WMD 7 ml, 95% CI -18.7 to 33.8) (Figure 6). Furthermore, the association between risk of requiring allogeneic blood transfusion and gelatin, when compared to crystalloids, became statistically insignificant when analysed by a random-effects model. This suggests that the adverse association between transfusion and gelatin may well be spurious. However, given the known risks associated with blood transfusion and the vast number of perioperative and critically ill patients who are treated with gelatin solutions on a daily basis, it is important for us to confirm the safety of gelatin with an adequately powered RCT. Cost-effective analysis should also be considered in such RCTs because gelatin solutions are more expensive than crystalloid solutions.

Our systematic review has some limitations. First, the majority of studies identified involved fewer than 100 patients and only two studies recruited more than 200 participants. The sample size in each stratum of comparison between gelatin and other types of intravenous fluid was also relatively small; hence the results were imprecise and prone to type II errors. The sample size of this meta-analysis on the mortality outcome (n=1900 patients) has a power of only 80% to detect a 5% increase in mortality compared to other types of intravenous fluid if the mortality rate of the latter group was 13%. Second, the patients included in the pooled trials were heterogeneous, ranging from major elective surgery (e.g. cardiac surgery, orthopaedic surgery or major gastrointestinal surgery) to critically ill patients with sepsis and hypovolaemia. Finally, most of the studies were designed to examine short-term outcomes only. The quality of the published studies on gelatin solutions was also unsatisfactory, with no study having both adequate allocation concealment and double-blinding. This could have introduced a potential bias due to the investigators' prior beliefs on the benefits and risks of gelatin solutions.

CONCLUSION

The current limited evidence suggests that using gelatin solutions compared to other fluids is not associated with either a beneficial or an adverse effect on mortality in critically ill patients. Moreover, using gelatin solutions as a plasma expander appears to have no significant advantage over crystalloids or isotonic albumin. However, they appear to be associated with a lower risk of acute renal failure compared to older starches. On the other hand, they may be associated with a slightly higher risk of requiring allogeneic blood transfusion for perioperative and critically ill patients. An adequately powered RCT with economic analysis is needed before gelatin solutions can be recommended as a routine plasma expander for patients undergoing major surgery or who are critically ill.

ACKNOWLEDGEMENTS

The authors would like to gratefully acknowledge Dr A. B. J. Groeneveld (Amsterdam, Netherlands), Dr T. Niemi (Helsinki, Finland), Dr P. Van der Linden (Brussels, Belgium) and Dr A. Schramko (Helsinki, Finland) for their assistance in providing original data from their studies for this meta-analysis.

FUNDING

This work was supported by the Department of Intensive Care at Royal Perth Hospital, Perth, Western Australia.

REFERENCES

(1.) The SAFE Study Investigators. A comparison of albumin and saline for fluid resuscitation in the intensive care unit. N Engl J Med 2004; 350:2247-2256.

(2.) Hogan J. The intravenous use of colloidal (gelatine) solutions in shock. JAMA 1915; 64:721-726.

(3.) Jacobson S, Smyth C. Gelatin as a substitute for plasma, observations on its administration to human beings. Arch Intern Med 1944; 74:254-257.

(4.) Ring J, Messmer K. Incidence and severity of anaphylactoid reactions to colloid volume substitutes. Lancet 1977; 1:466-469.

(5.) Mardel SN, Saunders FM, Allen H, Menezes G, Edwards CM, Ollerenshaw L et al. Reduced quality of clot formation with gelatin-based plasma substitutes. Br J Anaesth 1998; 80:204-207.

(6.) Evans PA, Glenn JR, Heptinstall S, Madira W. Effects of gelatin-based resuscitation fluids on platelet aggregation. Br J Anaesth 1998; 81:198-202.

(7.) Beyer R, Harmening U, Rittmeyer O, Zielmann S, Mielck F, Kazmaier S et al. Use of modified fluid gelatin and hydroxyethyl starch for colloidal volume replacement in major orthopaedic surgery. Br J Anaesth 1997; 78:44-50.

(8.) Niemi TT, Suojaranta-Ylinen RT, Kukkonen SI, Kuitunen AH. Gelatin and hydroxyethyl starch, but not albumin, impair hemostasis after cardiac surgery. Anesth Analg 2006; 102:998-1006.

(9.) Schramko A, Suojaranta-Ylinen R, Kuitunen A, Raivio P, Kukkonen S, Niemi T. Hydroxyethylstarch and gelatin solutions impair blood coagulation after cardiac surgery: a prospective randomized trial. Br J Anaesth 2010; 104:691-697.

(10.) Kuitunen A, Suojaranta-Ylinen R, Kukkonen S, Niemi T. A comparison of the haemodynamic effects of 4% succinylated gelatin, 6% hydroxyethyl starch (200/0.5) and 4% human albumin after cardiac surgery. Scand J Surg 2007; 96:72-78.

(11.) Schramko AA, Suojaranta-Ylinen RT, Kuitunen AH, Raivio PM, Kukkonen SI, Niemi TT. Comparison of the effect of 6% hydroxyethyl starch and gelatine on cardiac and stroke volume index: a randomized, controlled trial after cardiac surgery. Perfusion 2010; 25:283-291.

(12.) Cochrane Information Management System Review Manager Software. From http://ims.cochrane.org/revman/download/revman-4 Accessed March 2011. Anaesthesia and Intensive Care, Vol. 40, No. 1, January 2012

(13.) Asfar P, Kerkeni N, Labadie F, Gouello JP, Brenet O, Alquier P. Assessment of hemodynamic and gastric mucosal acidosis with modified fluid versus 6% hydroxyethyl starch: a prospective, randomized study. Intensive Care Med 2000; 26:1282-1287.

(14.) Beards SC, Watt T, Edwards JD, Nightingale P, Farragher EB. Comparison of the hemodynamic and oxygen transport responses to modified fluid gelatin and hetastarch in critically ill patients: a prospective, randomized trial. Crit Care Med 1994; 22:600-605.

(15.) Boks RH, Wijers MJ, Hofland J, Takkenberg JJ, Bogers AJ. Low molecular starch versus gelatin plasma expander during CPB: does it make a difference? Perfusion 2007; 22:333-337.

(16.) Gondos T, Marjanek Z, Ulakcsai Z, Szabo Z, Bogar L, Karolyi M et al. Short-term effectiveness of different volume replacement therapies in postoperative hypovolaemic patients. Eur J Anaesthesiol 2010; 27:794-800.

(17.) Inal MT, Memis D, Karamanlioglu B, Sut N. Effects of polygeline and hydroxyethyl starch solutions on liver functions assessed with LIMON in hypovolemic patients. J Crit Care 2010; 25:361.

(18.) Innerhofer P, Fries D, Margreiter J, Klingler A, Kuhbacher G, Wachter B et al. The effects of perioperatively administered colloids and crystalloids on primary platelet-mediated hemostasis and clot formation. Anesth Analg 2002; 95:858-865.

(19.) Jin S-L, Yu B-W. Effects of acute hypervolemic fluid infusion of hydroxyethyl starch and gelatin on hemostasis and possible mechanisms. Clin Appl Thromb Hemost 2010; 16:91-98.

(20.) Karoutsos S, Nathan N, Lahrimi A, Grouille D, Feiss P, Cox DJ. Thrombelastogram reveals hypercoagulability after administration of gelatin solution. Br J Anaesth 1999; 82:175-177.

(21.) Liang H, Yang C-X, Li H, Liu H-Z, Wang H-B. Effects of preoperative acute hypervolemic hemodilution on hypercoagulability of patients with colon cancer. Ai Zheng 2006; 25:12561260.

(22.) Mittermayr M, Streif W, Haas T, Fries D, Velik-Salchner C, Klingler A et al. Hemostatic changes after crystalloid or colloid fluid administration during major orthopedic surgery: the role of fibrinogen administration. Anesth Analg 2007; 105:905-917.

(23.) Molnar Z, Mikor A, Leiner T, Szakmany T. Fluid resuscitation with colloids of different molecular weight in septic shock. Intensive Care Med 2004; 30:1356-1360.

(24.) Mortelmans YJ, Vermaut G, Verbruggen AM, Arnout JM, Vermylen J, Van Aken H et al. Effects of 6% hydroxyethyl starch and 3% modified fluid gelatin on intravascular volume and coagulation during intraoperative hemodilution. Anesth Analg 1995; 81:1235-1242.

(25.) Ooi JS, Ramzisham AR, Zamrin MD. Is 6% hydroxyethyl starch 130/0.4 safe in coronary artery bypass graft surgery? Asian Cardiovasc Thorac Ann 2009; 17:368-372.

(26.) Parker MJ, Griffiths R, Boyle A. Preoperative saline versus gelatin for hip fracture patients; a randomized trial of 396 patients. Br J Anaesth 2004; 92:67-70.

(27.) Schortgen F, Lacherade JC, Bruneel F, Cattaneo I, Hemery F, Lemaire F et al. Effects of hydroxyethylstarch and gelatin on renal function in severe sepsis: a multicentre randomised study. Lancet 2001; 357:911-916.

(28.) Soares RR, Ferber L, Lorentz MN, Soldati MT. Intraoperative volume replacement: crystalloids versus colloids in surgical myocardial revascularization without cardiopulmonary bypass. Rev Bras Anestesiol 2009; 59:439-451.

(29.) Stockwell MA, Soni N, Riley B. Colloid solutions in the critically ill. A randomised comparison of albumin and polygeline. 1. Outcome and duration of stay in the intensive care unit. Anaesthesia 1992; 47:3-6.

(30.) Stockwell MA, Scott A, Day A, Riley B, Soni N. Colloid solutions in the critically ill. A randomised comparison of albumin and polygeline 2. Serum albumin concentration and incidences of pulmonary oedema and acute renal failure. Anaesthesia 1992; 47:7-9.

(31.) Tollofsrud S, Svennevig JL, Breivik H, Kongsgaard U, Ozer M, Hysing E et al. Fluid balance and pulmonary functions during and after coronary artery bypass surgery: Ringer's acetate compared with dextran, polygeline, or albumin. Acta Anaesthesiol Scand 1995; 39:671-677.

(32.) Trekova NA, Aksel'rod BA, Babalian GV, Zaitseva SV, Dement'eva II, Buniatian AA. Clinical and laboratory evaluation of 20% albumin (plasbumin) used in cardiosurgical patients. Anesteziol Reanimatol 2006; 2:18-21.

(33.) Upadhyay M, Singhi S, Murlidharan J, Kaur N, Majumdar S. Randomized evaluation of fluid resuscitation with crystalloid (saline) and colloid (polymer from degraded gelatin in saline) in pediatric septic shock. Indian Pediatr 2005; 42:223-231.

(34.) Van der Linden PJ, De Hert SG, Daper A, Trenchant A, Schmartz D, Defrance P et al. 3.5% urea-linked gelatin is as effective as 6% HES 200/0.5 for volume management in cardiac surgery patients. Can J Anesth 2004; 51:236-241.

(35.) Van der Linden PJ, De Hert SG, Deraedt D, Cromheecke S, De Decker K, De Paep R et al. Hydroxyethyl starch 130/0.4 versus modified fluid gelatin for volume expansion in cardiac surgery patients: the effects on perioperative bleeding and transfusion needs. Anesth Analg 2005; 101:629-634.

(36.) Vanhoonacker J, Ongenae M, Vanoverschelde H, Donadoni R. Hydroxyethyl starch 130/0.4 versus modified fluid gelatin for cardiopulmonary bypass priming: the effects on postoperative bleeding and volume expansion needs after elective CABG. Acta Anaesthesiol Belg 2009; 60:91-97.

(37.) Wahba A, Sendtner E, Strotzer M, Wild K, Birnbaum DE. Fluid therapy with Ringer's solution versus Haemaccel following coronary artery bypass surgery. Acta Anaesthesiol Scand 1996; 40:1227-1233.

(38.) Wahba A, Sendtner E, Birnbaum DE. Fluid resuscitation with Haemaccel vs. human albumin following coronary artery bypass grafting. Thorac Cardiovasc Surg 1996; 44:178-182.

(39.) Wu JJ, Huang MS, Tang GJ, Kao WF, Shih HC, Su CH et al. Hemodynamic response of modified fluid gelatin compared with lactated ringer's solution for volume expansion in emergency resuscitation of hypovolemic shock patients: preliminary report of a prospective, randomized trial. World J Surg 2001; 25:598-602.

(40.) Yap WW, Young D, Pathi V. Effects of gelatine and medium molecular weight starch as priming fluid in cardiopulmonary bypass--a randomised controlled trial. Perfusion 2007; 22:57-61.

(41.) American Thoracic Society. Evidence-based colloid use in the critically ill: American Thoracic Society Consensus Statement. Am J Respir Crit Care Med 2004; 170:1247-1259.

(42.) Egli GA, Zollinger A, Seifert B, Popovic D, Pasch T, Spahn DR. Effect of progressive haemodilution with hydroxyethyl starch, gelatin and albumin on blood coagulation. Br J Anaesth 1997; 78:684-689.

M. M. SAW *, B. CHANDLER ([dagger]), K. M. HO ([double dagger])

Intensive Care Department, Royal Perth Hospital, Perth, Western Australia, Australia

* MB, BS, BMedSci (Hons), Registrar, Department of Anaesthesia, Fremantle Hospital.

([dagger]) MB, BS, Anaesthetic Registrar, Anaesthetic Department, Royal Perth Hospital.

([double dagger]) MB, BS, PhD, FRCP, MPH, FCICM, FANZCA, Consultant Intensivist, Intensive Care Department, Royal Perth Hospital, School of Population Health, University of Western Australia.

Address for correspondence: Dr M. Saw. Email: melaniesaw@hotmail. com

Accepted for publication on September 26, 2011.
TABLE 1
Characteristics of included trials

Study, year, country Participants, Intervention, ml
of origin, funding inclusion and (volumes are mean
 exclusion criteria [+ or -] SD unless
 otherwise specified)

Asfar et al, 2000 Septic ventilated Colloid bolus given
(13), France patients with then haemodynamic and
 pulmonary artery tonometry measurements
 catheter and gastric taken at baseline and
 tonometer. Exclusions: at 60 min. Starch
 severe instability. group (n=16): 6% HES
 Mean age 64.5 y. 200/0.6 500. Gelatin
 group (n=18): 4%
 succinylated gelatin
 500.

Beards et al, 1994 Hypovolaemic ICU Colloid bolus given
(14), United Kingdom patients, mechanically prior to cardiac
 ventilated, with output measurements,
 pulmonary artery then haemodynamic
 catheter and femoral parameters measured
 arterial line. over 30 min. Starch
 Exclusions: active group (n = 13): HES
 bleeding. Mean age 55 450/0.7 500. Gelatin
 y. group (n = 15): 4%
 succinylated gelatin
 500.

Beyer et al, 1997 Patients undergoing All patients received
(7), Germany elective hip surgery Ringer's acetate as
 with expected blood baseline infusion with
 loss of more than 11. further colloid
 Exclusions: boluses given
 Haemoglobin less than according to
 11 g/ dl, ischaemic haemodynamic
 heart disease, parameters. Starch
 congestive cardiac group (n=19): 6% HES
 failure, clotting 200/0.5. Median=2500
 abnormalities, renal (90% CI 1500-2500).
 failure, pregnancy, Gelatin group (n=22):
 anticoagulant 3% gelatin.
 medication. Mean age Median=2400 (90% CI
 58.5 y. 2000-3000).

Boks et al, 2007 180 elective cardiac Bypass circuit
(15), Netherlands surgical patients priming. Starch group
 (valve replacement or (n=90): 6% HES 130-
 coronary artery bypass 0.4 1000 and 300-500
 grafting). Exclusions: Ringer's lactate.
 age more than 75 y, Gelatin group (n=90):
 preoperative 4% succinylated
 haemoglobin less than gelatin 300-1500.
 6 g/dl, body surface
 area less than 1.8
 [m.sup.2]. Mean age
 63.5 y.

Gondos et al, 2010 Postoperative 10 ml/kg fluid bolus
(16), Hungary. hypovolaemic patients. given over 30 min,
Fresenius-Kabi Exclusions: age less then haemodynamic
Hungary Inc., than 18 y, bleeding, parameters recorded
Budapest, Hungary; shock, severe over following 2 h.
Pulsion Medical pulmonary oedema, Starch group (n=50):
Systems AG, Munich, uraemia, allergy to 6% HES 130/0.4.
Germany; MEDIAL colloids, life Gelatin group (n=50):
Inc., Budapest, expectancy less than 4% succinylated
Hungary; HUMAN 24 h, trauma. Mean age gelatin. Albumin group
BioPlazma LLC, 57.5 y. (n=50): 5% human
Godollo, Hungary albumin. Crystalloid
 group (n=50): Lactated
 Ringer's solution.

Inal et al, 2010 Adult ICU patients. Colloid bolus prior to
(17), Turkey. Exclusions: pregnancy, LIMON measurements
University of Trakya cardiovascular (non-invasive liver
research grant instability, heart function monitoring
 failure, renal system). Gelatin group
 failure, liver (n=15): 3.5%
 failure, brain death. polygeline 500. Starch
 group (n=15): 6% HES
 130/0.5 500.

Innerhoffer et al, Patients undergoing Baseline Ringer's
2002 (18), Austria. primary total knee lactate infusion given
Fresenius GmbH replacement. in all groups. Starch
Austria; B. Braun, Exclusions: any group (n=20): HES
Melsungen, contraindication to 200/0.5 3 ml/kg/h and
Germany regional block or 1:1 replacement for
 arterial puncture, blood loss 1242 [+ or
 primary or secondary -] 315. Gelatin group
 bleeding disorders, (n=20): 4%
 hepatic or renal succinylated gelatin 4
 disorders, platelet ml/kg/h 1:1.3
 inhibiting drugs. Mean replacement for blood
 age 69 y. loss 1435 [+ or -]
 469. Crystalloid group
 (n=20): Ringer's
 lactate 10 ml/kg/h and
 1:3 replacement for
 blood loss 4801 [+ or
 -] 1239.

Jin et al, 2010 36 patients with Acute hypervolaemic
(19), China. gastric cancer infusion: 30 ml/kg/h
Shanghai Science undergoing from 20 min prior to
and Technology gastrectomy. surgery to 40 min
Development Fund Exclusion: cardiac after induction of
 insufficiency (New general anaesthesia.
 York Heart Association Gelatin group (n=12):
 class III to IV), 4% succinylated
 renal insufficiency, gelatin 3809 [+ or -]
 altered liver 392. Crystalloid group
 function, preoperative (n=12): Ringer's
 anaemia, preoperative lactate 4190 [+ or -]
 coagulation 327. HES group (n=12):
 abnormalities, known HES 130/0.4 3916 [+ or
 hypersensitivity to -] 666.
 study fluids, use of
 anticoagulant or
 antiplatelet agents
 prior to surgery. Mean
 age 52.3 y.

Karoutsos et al, Patients undergoing Intraoperative volume
1999 (20), France elective total knee or replacement. Starch
 hip replacement group (n=15): 6% HES
 surgery. Exclusions: 200/0.62 1:1
 abnormal clotting, low replacement for blood
 platelets, medications loss 1450 [+ or -]
 impairing haemostasis, 288. Gelatin group
 renal failure, allergy (n=15): 3.5% modified
 to plasma substitutes. gelatin 1:1.5
 Mean age 63.6 y. replacement for blood
 loss 1800 [+ or -]
 538. Albumin group (n
 = 12): Albumin 5% 1:1
 replacement for blood
 loss 1366 [+ or -]
 255.

Liang et al, 2006 60 patients with colon Acute hypervolaemic
(21), China cancer undergoing haemodilution: 15
 surgery. ml/kg given 30 min
 preoperatively. Starch
 group (n=20): 6% HES
 1135 [+ or -] 162.
 Gelatin group (n=20):
 4% succinylated
 gelatin 1195 [+ or -]
 151. Crystalloid group
 (n=20): Ringer's
 lactate 2887 [+ or -]
 231.

Mittermayr et al, Spinal surgery of All patient received
2007 (22), Austria. duration more than 3 baseline Ringer's
Fresenius, Pharma h. Exclusions: Allergy solution infusion.
Austria GmbH, Graz, to colloid, Starch group (n = 19):
Austria; B. Braun, haemostatic problems. 6% HES 130/0.4 6 to 8
Maria Enzersdorf, Mean age 48 y. ml/kg/h. Median=894
Austria (range 546-1300).
 Gelatin group (n=21):
 4% succinylated
 gelatin 8-11 ml/kg/h.
 Median=928 (range
 683-1250). Crystalloid
 group (n=20): Lactated
 Ringer's solution
 13-15 ml/kg/h.
 Median=3650 (range
 1700-5500).

Molnar et al, 2004 Mechanically 250 ml boluses to max
(23), Hungary. ventilated ICU of 1000 ml then
National research patients with septic haemodynamic
body funding shock. Exclusions: measurements. Starch
 heart failure, chronic group (n=15): 6% HES
 respiratory failure, 200/0.6. Gelatin group
 chronic renal failure (n=15): 4%
 requiring renal succinylated gelatin.
 replacement therapy,
 chronic liver failure,
 diabetes mellitus,
 known aortic aneurysm,
 chronic liver failure.
 Mean age 52.5 y.

Mortelmans et al, Patients undergoing All patients received
1995 (24), Belgium primary total hip baseline crystalloid
 replacement surgery. infusion. 1000 ml of
 Exclusions: randomised colloid was
 haematocrit less than given prior to surgery
 0.35, coagulopathy, for acute
 antiplatelet agents, haemodilution, and
 renal impairment, further 250 ml boluses
 allergy to artificial if needed
 colloids. Mean age intraoperatively.
 61.5 y. Starch group (n=21):
 HES 200/0.4 2119 [+ or
 -] 350. Gelatin group
 (n=21): 3% gelatin
 2131 [+ or -] 516.

Niemi et al, 20068, 45 elective cardiac Volume replacement (15
Finland. surgical patients. ml-kg) given
Department of Exclusions: postoperatively on
Anesthesiology preoperative arrival to ICU, rate
and Intensive Care coagulation disorders, titrated to
Medicine, Helsinki renal or hepatic haemodynamic
University Hospital failure, patients on parameters. Albumin
 anticoagulant group (n=15): 4%
 medication or albumin. Median=1125
 antiplatelet therapy. (range 750-1500).
 Mean age 64 y. Starch group (n=15):
 6% HES 200-0.5.
 Median=1185 (range
 750-1530). Gelatin
 group (n=15): 4%
 succinylated gelatin.
 Median=1200 (range
 750-1650).

Ooi Su Min et al, 90 elective cardiac Colloid used as
2009 (25), Malaysia surgical patients cardiopulmonary bypass
 undergoing coronary circuit priming and
 artery bypass postoperative volume
 grafting. Exclusions: replacement titrated
 repeat coronary artery to CVP. Starch group
 bypass grafting, (n=45): 6% HES 130-
 congestive heart 0.4 1942 [+ or -]
 failure, recent 1046. Gelatin group
 antiplatelet therapy, (n=45): 4%
 coagulopathy, renal succinylated gelatin
 dysfunction, liver 1973 [+ or -] 729.
 dysfunction, history
 of pancreatitis, known
 hypersensitivity. Mean
 age 58 y.

Parker et al, 2004 Patients with 500 ml of study fluid
(26), United fractured neck of given preoperatively.
Kingdom. femur for surgery. Gelatin group (n=198):
Braun Medical Exclusions: 4% succinylated
 conservative gelatin. Crystalloid
 management of group (n=198): 0.9%
 fracture, congestive saline.
 cardiac failure. Mean
 age 80 y.

Schortgen et al, Intensive care 500 ml colloid boluses
2001 (27), France patients with septic as required by
 shock and physician. Starch
 hypovolaemia. group (n=65): 6% HES
 Exclusions: pregnancy, 200/0.6. Median=31
 allergy to HES or ml/kg (IQR 19-51).
 gelatin, severe renal Gelatin group (n=64):
 dysfunction, prior 3% gelatin. Median=43
 administration of HES ml/kg (IQR 19-60).
 or mannitol. Median
 ages: 56 y in gelatin
 group, 60 y HES group.

Schramko et al, 45 elective cardiac Postoperative volume
2010 (9), Finland. surgical patients. replacement, total
Government Grant Exclusions: dose 28 ml/kg over 16
for Healthcare preoperative to 20 h. Starch group
Research, Finland coagulation disorder, (n=15): 6% HES 130/0.4
 renal or hepatic 28 ml/kg. Crystalloid
 failure, preoperative group (n=15): Ringer's
 left ventricular acetate. Gelatin group
 ejection fraction less (n=15): 4%
 than 40%, succinylated gelatin
 antithrombotic or 28 ml/kg.
 antiplatelet therapy.
 Mean age 64 y.

Soares et al, 2009 40 elective cardiac Intraoperative fluid
(28), Brazil surgical patients replacement, titrated
 undergoing off-pump to CVP >10 mmHg.
 coronary Gelatin group (n=20):
 revascularisation. Modified fluid gelatin
 Exclusions: bleeding 15 ml/kg. Crystalloid
 disorders, myocardial group (n=20): Normal
 infarction within saline.
 three weeks,
 haemoglobin below 10
 g/dl, renal or liver
 failure, anticoagulant
 medication or
 emergency surgery.
 Mean age 66.5 y.

Stockwell et al, Intensive care unit Fluids given as per
1992 (29,30), United patients aged more clinical or
Kingdom than 18 y. Exclusions: haemodynamic
 admission for cardiac parameters, baseline
 monitoring or crystalloid infusion.
 thrombolysis. Mean age Gelatin group (n=249):
 62 y. 3.5% polygeline 3060
 (range 0-30000).
 Albumin Group (n=226):
 4.5% human albumin
 2500 (range 0-30000).

Tollofsrud et al, 40 elective cardiac Intraoperative and
1995 (31), Norway surgical patients postoperative volume
 undergoing coronary replacement for 48 h.
 artery bypass Crystalloid group
 grafting. Exclusions: (n=10): Ringer's
 left ventricular acetate 1761 [+ or -]
 ejection fraction less 1726. Gelatin group
 than 40%, valvular (n=10): 3.5%
 heart disease, polygeline 2100 [+ or
 ventricular aneurysm, -] 1125. Dextran group
 significant (n=10): Dextran 70
 arrhythmia, diabetes 1055 [+ or -] 760.
 mellitus, renal Albumin group (n=10):
 failure or lung 4% Albumin 975 [+ or
 disease. Mean age 64.5 -] 322.
 y.

Trekova et al, 2006 36 elective cardiac Bypass circuit
(32), Russia surgical patients priming. Gelatin group
 undergoing coronary (n=18): 4%
 artery bypass grafting succinylated gelatin
 or valvular surgery. 500. Albumin group (n
 Age ranges: 6 to 74 y = 18): 20% albumin 150
 in albumin group, 16 or 200.
 to 65 y in gelatin
 group.

Upadhyay et al, Paediatric patients 20 ml/kg fluid
2004 (33), India aged one month to 12 y boluses. Gelatin group
 with septic shock. (n=29): 3.5%
 Exclusions: polygeline median=30
 multi-organ failure, ml/kg (90% CI 20-70).
 immunosuppressed Crystalloid group
 treatment or disease. (n=31): 0.9% Saline 50
 Mean age 33 months. ml/kg (90% CI 20-108).

Van der Linden et 110 elective cardiac Bypass circuit priming
al, 2004 (34), surgical patients and perioperative
Belgium. Fresenius undergoing coronary volume replacement.
Kabi GmbH, Bad artery bypass grafting Starch group (n=55):
Homburg, Germany and/or single valve 6% HES 200/0.5 1813 [+
 surgery. Exclusions: or -] 444. Gelatin
 Left ventricular group (n=55): 3.5%
 ejection fraction less polygeline 1961 [+ or
 than 35%, combined -] 365.
 cardiac surgery, redo
 operations, known
 hypersensitivity,
 liver or renal
 dysfunction. Mean age
 63 y.

Van der Linden et 132 elective cardiac Bypass circuit priming
al, 2005 (35), surgical patients and perioperative
Belgium undergoing coronary volume replacement
 artery bypass (maximum 50 ml/kg/d).
 grafting. Exclusions: Gelatin group (n=68):
 left ventricular 3% modified fluid
 ejection fraction less gelatin 3814 [+ or -]
 than 35%, known 1139. Starch group
 hypersensitivity, (n=64):
 liver or renal 6% HES
 130/0.4 3863 [+ or -]
 1359. dysfunction.
 Mean age 66.5 y.

Vanhoonacker et al, 157 elective cardiac Bypass circuit
2009 (36), Belgium surgical patients priming. Starch group
 undergoing coronary (n=82): 6% HES 130/
 artery bypass 0.4 1500. Gelatin
 grafting. Exclusions: group (n=72): 3%
 left ventricular succinylated gelatin
 ejection fraction less 1500.
 than 40%, liver or
 renal dysfunction.
 Mean age 66.5 y.

Wahba et al, 1996 22 elective cardiac Postoperative volume
(37), Germany surgical patients replacement titrated
 undergoing coronary to haemodynamic
 artery bypass parameters.
 grafting. Exclusions: Crystalloid group
 left ventricular (n=10): Ringer's
 dysfunction, solution 13 ml/kg/h
 antiplatelet therapy, infusion and 500 ml
 intravenous heparin, boluses. Gelatin group
 clotting disorder. (n=10): 3.5%
 Mean age 64.3 y. polygeline 6 ml/kg/h
 infusion and 200 ml
 boluses.

Wahba et al, 1996 20 elective cardiac Postoperative volume
(38), Germany surgical patients replacement given over
 undergoing coronary 8 h, titated to mean
 artery bypass arterial pressure and
 grafting. Exclusions: pulmonary capillary
 left ventricular wedge pressure.
 dysfunction, Albumin group (n=10):
 antiplatelet therapy, 5% albumin 3997 [+ or
 intravenous heparin, -] 647. Gelatin group
 clotting disorder. (n=10): 3.5%
 Mean age 64.5 y. polygeline 4010 [+ or
 -] 530.

Wu et al, 2001 (39), Hypotensive trauma 1000 ml of fluid given
Taiwan patients aged more over 10 to 15 min and
 than 16 y. Exclusions: haemodynamic
 pregnancy, invasive parameters recorded
 ventilation, CCF, over the following
 refractory to filling. hour. All patients
 Mean age 44.6 y. also given 1000 ml of
 lactated Ringer's
 solution over 1 h.
 Gelatin group (n=18):
 Succinylated gelatin
 4%. Crystalloid group
 (n=16): lactated
 Ringer's solution.

Yap et al, 2007 40 elective cardiac Bypass circuit
(40), United Kingdom surgical patients priming. Starch group
 undergoing coronary (n=20): 6% HES 130/0.4
 artery bypass 500. Gelatin group
 grafting. Exclusions: (n=20): 4%
 age less than 45 or succinylated gelatin
 more than 75 y, 500.
 emergency operations,
 anaemia, coagulation
 disorders, pregnancy,
 hepatic or renal
 dysfunction, known
 hypersensitivity,
 aprotinin infusion
 intraoperatively and
 history of glaucoma.
 Mean age 61 y.

Study, year, country Outcomes and Allocation
of origin, funding study period concealment, blinding,
 intention to treat
 analysis, proportion
 completing study

Asfar et al, 2000 Mortality Unclear allocation
(13), France concealment. Single-
 centre open RCT. Not
 intention to treat
 analysis. 94% (34/36)
 of participants
 completed the study.

Beards et al, 1994 Mortality Unclear allocation
(14), United Kingdom concealment. Single-
 centre open RCT.
 Intention to treat
 analysis used. 100%
 (28/28) completed
 study.

Beyer et al, 1997 Allergic reactions. Unclear allocation
(7), Germany Amount of blood concealment. Single-
 transfused. Followed- centre single-blinded
 up for 24 h. RCT. Not intention to
 treat analysis. 85%
 (41/48) of
 participants completed
 study.

Boks et al, 2007 Allergic reactions. Allocation concealment
(15), Netherlands Total blood loss. inadequate. Single-
 Amount of blood centre open RCT.
 transfused. Followed Intention to treat
 until ICU discharge. analysis used. All
 participants completed
 the study.

Gondos et al, 2010 Mortality Allocation concealment
(16), Hungary. inadequate.
Fresenius-Kabi Multicentre (11 ICUs)
Hungary Inc., open RCT. Intention to
Budapest, Hungary; treat analysis used.
Pulsion Medical All participants
Systems AG, Munich, completed the study.
Germany; MEDIAL
Inc., Budapest,
Hungary; HUMAN
BioPlazma LLC,
Godollo, Hungary

Inal et al, 2010 Mortality Unclear allocation
(17), Turkey. concealment. Single-
University of Trakya centre open RCT.
research grant Intention to treat
 analysis used. All
 participants completed
 the study.

Innerhoffer et al, Total blood loss. Allocation concealment
2002 (18), Austria. Amount of blood adequate. Single-
Fresenius GmbH transfused. Proportion centre open RCT.
Austria; B. Braun, of patients Intention to treat
Melsungen, transfused. Followed analysis used. All
Germany until 2 h participants completed
 postoperatively. the study.

Jin et al, 2010 Total blood loss. Allocation concealment
(19), China. Followed until 4 h inadequate. Single-
Shanghai Science after end of infusion. centre double-blinded
and Technology RCT. Intention to
Development Fund treat analysis used.
 All participants
 completed the study.

Karoutsos et al, Total blood loss. Allocation concealment
1999 (20), France Amount of blood unclear. Single-
 transfused. Followed centre open RCT.
 until 24 h Intention to treat
 postoperatively. analysis used All
 participants completed
 the study.

Liang et al, 2006 Total blood loss. Allocation concealment
(21), China Followed until 1 h adequate. Single-
 postoperatively. centre RCT. Intention
 to treat analysis
 used. All participants
 completed the study.

Mittermayr et al, Total blood loss. Adequate allocation
2007 (22), Austria. Proportion of patients concealment. Single-
Fresenius, Pharma transfused. Followed centre open RCT. Not
Austria GmbH, Graz, up until day 5 intention to treat
Austria; B. Braun, postoperatively. analysis. 91% (60/66)
Maria Enzersdorf, of participants
Austria completed the study.

Molnar et al, 2004 Mortality Allocation concealment
(23), Hungary. unclear. Single-
National research centre open RCT.
body funding Intention to treat
 analysis used. All
 participants completed
 the study.

Mortelmans et al, Adverse drug Allocation concealment
1995 (24), Belgium reactions. Total blood unclear. Single-
 loss. Amount of blood centre open RCT.
 transfused. Followed Intention to treat
 up until 3 h analysis used. All
 postoperatively. participants completed
 the study.

Niemi et al, 20068, Total blood loss. Allocation concealment
Finland. Proportion of patients inadequate. Single-
Department of transfused. Followed centre open RCT.
Anesthesiology until first Intention to treat
and Intensive Care postoperative morning. analysis used. All
Medicine, Helsinki participants completed
University Hospital the study.

Ooi Su Min et al, Mortality. Number of Allocation concealment
2009 (25), Malaysia patients with acute unclear. Single-
 renal failure *. centre single-blinded
 Allergic reactions. RCT. Intention to
 Total blood loss. treat analysis used.
 Proportion of patients All participants
 transfused. Hospital completed the study.
 length of stay.
 Followed for 24 h
 postoperatively.

Parker et al, 2004 Mortality. Adverse Allocation concealment
(26), United drug reactions. unclear. Single-
Kingdom. Proportion requiring centre double-blinded
Braun Medical transfusion. Duration RCT. Intention to
 of hospital stay. treat analysis used.
 Followed up until 1 y All participants
 postoperatively. completed the study.

Schortgen et al, Mortality. Number of Adequate allocation
2001 (27), France patients with acute concealment.
 renal failure. Multicentre open RCT.
 Intention to treat
 analysis used. All
 participants completed
 the study.

Schramko et al, Total blood loss. Allocation concealment
2010 (9), Finland. Proportion of patients inadequate. Single-
Government Grant transfused. Amount of centre open RCT.
for Healthcare blood transfused. Intention to treat
Research, Finland Followed until first analysis used. All
 postoperative morning. participants completed
 the study.

Soares et al, 2009 Mortality. Number of Allocation concealment
(28), Brazil patients with acute inadequate. Single-
 renal failure (#). centre open RCT.
 Total blood loss. Intention to treat
 Proportion of patients analysis used. All
 transfused. Amount of participants completed
 blood transfused. the study.
 Hospital length of
 stay. Followed until
 hospital discharge.

Stockwell et al, Mortality. Number of Allocation concealment
1992 (29,30), United patients with acute unclear. Single-
Kingdom renal failure. centre open RCT.
 Followed up until ICU Intention to treat
 discharge. analysis not used. All
 participants completed
 the study.

Tollofsrud et al, Total blood loss. Allocation concealment
1995 (31), Norway Total amount of blood inadequate. Single-
 transfused. Followed centre open RCT.
 until 48 h Intention to treat
 postoperatively. analysis used. All
 participants completed
 the study.

Trekova et al, 2006 Total amount of blood Allocation concealment
(32), Russia loss. Followed until 3 unclear. Single-
 h postoperatively. centre open RCT.
 Intention to treat
 analysis used. All
 patients completed the
 study.

Upadhyay et al, Mortality. Number of Allocation concealment
2004 (33), India patients with acute unclear. Single-
 renal failure. centre open RCT.
 Intention to treat
 analysis used. All
 participants completed
 the study.

Van der Linden et Hospital mortality. Allocation concealment
al, 2004 (34), Allergic reactions. inadequate. Single-
Belgium. Fresenius Total blood loss. centre open RCT.
Kabi GmbH, Bad Proportion of patients Intention to treat
Homburg, Germany transfused. Followed analysis used. All
 for 18 h participants completed
 postoperatively (blood the study.
 loss).

Van der Linden et Hospital mortality. Allocation concealment
al, 2005 (35), Total blood loss. adequate. Single-
Belgium Proportion of patients centre open RCT.
 transfused. Amount of Intention to treat
 blood transfused. analysis used. All
 Hospital length of participants completed
 stay. Followed for 20 the study.
 h postoperatively
 (blood loss).

Vanhoonacker et al, Bypass circuit Allocation concealment
2009 (36), Belgium priming. Starch group unclear. Single/
 (n=82): 6% HES 130/ centre double/blinded
 0.4 1500. Gelatin RCT. Intention to
 group (n=72): 3% treat analysis not
 succinylated gelatin used. 98% (154/157) of
 1500. patients patients completed the
 transfused. Amount of study.
 blood transfused.
 Followed for 24 h
 postoperatively.

Wahba et al, 1996 Total blood loss Unclear allocation
(37), Germany Amount of blood concealment. Single/
 transfused. Followed centre open RCT.
 until first Intention to treat
 postoperative morning. analysis not used. 90%
 (20/22) of patients
 completed the study.

Wahba et al, 1996 Total blood loss. Allocation concealment
(38), Germany Followed until first unclear. Single-
 postoperative morning. centre open RCT.
 Intention to treat
 analysis used. All
 participants completed
 the study.

Wu et al, 2001 (39), Mortality. Adverse Allocation concealment
Taiwan drug reactions. unclear. Single/
 centre open RCT. 83%
 (34/41) of patients
 completed the study.

Yap et al, 2007 Total blood loss. Adequate allocation
(40), United Kingdom Followed for 14 h concealment. Single-
 postoperatively. centre open RCT.
 Intention to treat
 analysis used. All
 participants completed
 the study.

HES=hydroxyethyl starch, RCT=randomised controlled trial,
ICU=intensive care unit, CVP=central venous pressure,
IQR=interquartile range, CCF=congestive cardiac failure.
* Proportion of patients needing dialysis. # Defined as
creatinine increase of more than 1.5 mg/dl.

Figure 2: Mortality (all cause hospital or 28 day).

Outcome Mortality (all cause hospital or 28 day)

Study Gelatin Control Weight
 n/N n/N %

01 Gelatin vs Crystalloid
Wu 2001 (39) 2/18 3/16 2.39
Parker 2004 (26) 19/198 9/198 6.88
Upadhyay 2005 (33) 9/29 9/31 5.08
Soares 2009 (28) 0/20 1/20 1.24
Gondos 2010 (16) 12/50 14/50 9.00
Subtotal (95% CI) 315 315 24.59
Total events: 42 (Gelatin), 36 (Control)
Test for heterogeneity: [I.sup.2]=5.7%
 Test for overall effect: Z=0.78 (P=0.43)

02 Gelatin vs Albumin
Stockwell 1992 (29) 50/249 45/226 31.90
Gondos 2010 (16) 12/50 12/50 7.72
Subtotal (95% CI) 299 276 39.61
Total events: 62 (Gelatin), 57 (Control)
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=0.04 (P=0.97)

03 Gelatin vs Starch
Beards 1994 (14) 6/15 5/13 2.72
Asfar 2000 (13) 12/18 10/16 2.99
Schortgen 2001 (27) 33/64 34/65 13.82
Molnar 2004 (23) 10/15 12/15 3.38
Van der Linden 2004 (34) 0/55 0/55
Van der Linden 2005 (35) 1/1 0/64 0.43
Ooi Su Min 2009 (25) 0/45 0/45
Gondos 2010 (16) 12/50 15/50 9.64
Inal 2010 (17) 5/15 5/15 2.82
Subtotal (95% CI) 345 338 35.8
Total events: 79 (Gelatin), 81 (Control)
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=0.40 (P=0.69)

Total (95% CI) 959 929 100.00
Total events: 183 (Gelatin), 174 (Control)
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=0.20 (P=0.84)

Study OR (fixed)
 95% CI

01 Gelatin vs Crystalloid
Wu 2001 (39) 0.54 [0.08, 3.74]
Parker 2004 (26) 2.23 [0.98, 5.06]
Upadhyay 2005 (33) 1.10 [0.36, 3.32]
Soares 2009 (28) 0.32 [0.01, 8.26]
Gondos 2010 (16) 0.81 [0.33, 1.99]
Subtotal (95% CI) 1.22 [0.75, 1.99]
Total events: 42 (Gelatin), 36 (Control)
Test for heterogeneity: [I.sup.2]=5.7%
 Test for overall effect: Z=0.78 (P=0.43)

02 Gelatin vs Albumin
Stockwell 1992 (29) 1.01 [0.64, 1.59]
Gondos 2010 (16) 1.00 [0.40, 2.50]
Subtotal (95% CI) 1.01 [0.67, 1.51]
Total events: 62 (Gelatin), 57 (Control)
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=0.04 (P=0.97)

03 Gelatin vs Starch
Beards 1994 (14) 1.07 [0.23, 4.89]
Asfar 2000 (13) 1.20 [0.29, 4.91]
Schortgen 2001 (27) 0.97 [0.49, 1.94]
Molnar 2004 (23) 0.50 [0.10, 2.63]
Van der Linden 2004 (34) Not estimable
Van der Linden 2005 (35) 2.87 [0.11, 71.66]
Ooi Su Min 2009 (25) Not estimable
Gondos 2010 (16) 0.74 [0.30, 1.79]
Inal 2010 (17) 1.00 [0.22, 4.56]
Subtotal (95% CI) 0.91 [0.59, 1.41]
Total events: 79 (Gelatin), 81 (Control)
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=0.40 (P=0.69)

Total (95% CI) 1.03 [0.80, 1.32]
Total events: 183 (Gelatin), 174 (Control)
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=0.20 (P=0.84)

OR=odds ratio, CI=confidence interval.

FIGURE 3: Risk of renal failure.

Outcome: Acute Renal Failure

Study Gelatin Control Weight
 n/N n/N %

01 Gelatin vs Crystalloid
Upadhyay 2005 (33) 1/29 3/31 9.98
Soares 2009 (28) 1/20 0/20 1.65
Subtotal (95% CI) 49 51 11.63
Total events: 2 (Gelatin), 3 (Control)
Test for heterogeneity: [I.sup.2]=17.5%
Test for overall effect: Z=0.36 (P=0.72)

02 Gelatin vs Albumin
Stockwell 1992 (30) 11/59 5/53 15.27
Subtotal (95% CI) 59 53 15.27
Total events: 11 (Gelatin), 5 (Control)
Test for heterogeneity: not applicable
Test for overall effect: Z=1.37 (P=0.17)

03 Gelatin vs Starch
Schortgen 2001 (27) 15/64 27/65 73.09
Ooi Su Min 2009 (25) 0/45 0/45
Subtotal (95% CI) 4/18 110 73.09
Total events: 15 (Gelatin), 27 (Control)
Test for heterogeneity: not applicable
Test for overall effect: Z=2.17 (P=0.03)

Total (95% CI) 217 214 100
Total events: 28 (Gelatin), 35 (Control)
Test for heterogeneity: [I.sup.2]=55.4%
Test for overall effect: Z=1.06 (P=0.29)

Study OR (fixed)
 95% CI

01 Gelatin vs Crystalloid
Upadhyay 2005 (33) 0.33 [0.03, 3.40]
Soares 2009 (28) 3.15 [0.12, 82.16]
Subtotal (95% CI) 0.73 [0.14, 3.89]
Total events: 2 (Gelatin), 3 (Control)
Test for heterogeneity: [I.sup.2]=17.5%
Test for overall effect: Z=0.36 (P=0.72)

02 Gelatin vs Albumin
Stockwell 1992 (30) 2.20 [0.71, 6.81]
Subtotal (95% CI) 2.20 [0.71, 6.81]
Total events: 11 (Gelatin), 5 (Control)
Test for heterogeneity: not applicable
Test for overall effect: Z=1.37 (P=0.17)

03 Gelatin vs Starch
Schortgen 2001 (27) 0.43 [0.20, 0.92]
Ooi Su Min 2009 (25) Not estimable
Subtotal (95% CI) 0.43 [0.20, 0.92]
Total events: 15 (Gelatin), 27 (Control)
Test for heterogeneity: not applicable
Test for overall effect: Z=2.17 (P=0.03)

Total (95% CI) 0.74 [0.42, 1.29]
Total events: 28 (Gelatin), 35 (Control)
Test for heterogeneity: [I.sup.2]=55.4%
Test for overall effect: Z=1.06 (P=0.29)

OR=odds ratio, CI=confidence interval.

FIGURE 4: Amount of allogeneic blood transfusion in millilitres.

Outcome: Amount Blood Transfused

 Gelatin
Study N Mean (SD) N

01 Gelatin vs Crystalloid
Tollofsrud 1995 (31) 10 490.00 (548.00) 10
Wahba 1996 (37) 10 330.00 (390.00) 10
Innerhofer 2002 (18) 20 15.00 (75.00) 20
Soares 2009 (28) 20 120.00 (240.00) 20
Schramko 2010 (9) 15 420.00 (600.00) 15
Subtotal (95% CI) 75 75
Test for heterogeneity: [I.sub.2]=62.9%
Test for overall effect: Z=0.72 (P=0.47)

02 Gelatin vs Albumin
Tollofsrud 1995 (31) 10 490.00 (548.00) 10
Karoutsos 1999 (20) 15 342.00 (333.00) 12
Subtotal (95% CI) 25 22
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=2.05 (P=0.04)

03 Gelatin vs Starch
Mortelmans 1995 (24) 21 459.00 (225.00) 21
Beyer 1997 (7) 22 1250.00 (625.00) 19
Karoutsos 1999 (20) 15 342.00 (333.00) 15
Innerhofer 2002 (18) 20 15.00 (75.00) 20
Van der Linden 2005 (35) 68 0.00 (1.50) 64
Boks 2007 (15) 90 582.00 (541.00) 90
Vanhoonacker 2009 (36) 72 251.00 (366.00) 82
Schramko 2010 (9) 15 420.00 (600.00) 15
Subtotal (95% CI) 323 326
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=0.00 (P=1.00)

04 Gelatin vs Dextran
Tollofsrud 1995 (31) 10 490.00 (548.00) 10
Subtotal (95% CI) 10 10
Test for heterogeneity: not applicable
Test for overall effect: Z=0.46 (P=0.65)

Total (95% CI) 433 433
Test for heterogeneity: [I.sub.2]=28.6%
Test for overall effect: Z=0.00 (P=1.00)

 Control Weight
Study Mean (SD) %

01 Gelatin vs Crystalloid
Tollofsrud 1995 (31) 67.00 (200.00) 0.00
Wahba 1996 (37) 390.00 (510.00) 0.00
Innerhofer 2002 (18) 60.00 (150.00) 0.00
Soares 2009 (28) 300.00 (540.00) 0.00
Schramko 2010 (9) 159.00 (150.00) 0.00
Subtotal (95% CI) 0.01
Test for heterogeneity: [I.sub.2]=62.9%
Test for overall effect: Z=0.72 (P=0.47)

02 Gelatin vs Albumin
Tollofsrud 1995 (31) 240.00 (310.00) 0.00
Karoutsos 1999 (20) 178.00 (165.00) 0.00
Subtotal (95% CI) 0.00
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=2.05 (P=0.04)

03 Gelatin vs Starch
Mortelmans 1995 (24) 355.00 (236.00) 0.00
Beyer 1997 (7) 1200.00 (313.00) 0.00
Karoutsos 1999 (20) 285.00 (306.00) 0.00
Innerhofer 2002 (18) 60.00 (150.00) 0.00
Van der Linden 2005 (35) 0.00 (1.50) 99.98
Boks 2007 (15) 586.00 (522.00) 0.00
Vanhoonacker 2009 (36) 302.00 (380.00) 0.00
Schramko 2010 (9) 300.00 (300.00) 0.00
Subtotal (95% CI) 99.99
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=0.00 (P=1.00)

04 Gelatin vs Dextran
Tollofsrud 1995 (31) 390.00 (417.00) 0.00
Subtotal (95% CI) 0.00
Test for heterogeneity: not applicable
Test for overall effect: Z=0.46 (P=0.65)

Total (95% CI) 100.00
Test for heterogeneity: [I.sub.2]=28.6%
Test for overall effect: Z=0.00 (P=1.00)

 WMD (fixed)
Study 95% CI

01 Gelatin vs Crystalloid
Tollofsrud 1995 (31) 423.00 [61.44, 784.56]
Wahba 1996 (37) -60.00 [-457.93, 337.93]
Innerhofer 2002 (18) -45.00 [-118.50, 28.50]
Soares 2009 (28) -180.00 [-438.98, 78.98]
Schramko 2010 (9) 261.00 [-51.98, 573.98]
Subtotal (95% CI) -24.50 [-91.28, 42.29]
Test for heterogeneity: [I.sub.2]=62.9%
Test for overall effect: Z=0.72 (P=0.47)

02 Gelatin vs Albumin
Tollofsrud 1995 (31) 250.00 [-140.23, 640.23]
Karoutsos 1999 (20) 164.00 [-28.65, 356.65]
Subtotal (95% CI) 180.85 [8.11, 353.60]
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=2.05 (P=0.04)

03 Gelatin vs Starch
Mortelmans 1995 (24) 104.00 [-35.46, 243.46]
Beyer 1997 (7) 50.00 [-246.67, 346.67]
Karoutsos 1999 (20) 57.00 [-171.86, 285.86]
Innerhofer 2002 (18) -45.00 [-118.50, 28.50]
Van der Linden 2005 (35) 0.00 [-0.51, 0.51]
Boks 2007 (15) -4.00 [-159.32, 151.32]
Vanhoonacker 2009 (36) -51.00 [-168.95, 66.95]
Schramko 2010 (9) 120.00 [-219.48, 459.48]
Subtotal (95% CI) 0.00 [-0.51, 0.51]
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=0.00 (P=1.00)

04 Gelatin vs Dextran
Tollofsrud 1995 (31) 100.00 [-326.80, 526.80]
Subtotal (95% CI) 100.00 [-326.80, 526.80]
Test for heterogeneity: not applicable
Test for overall effect: Z=0.46 (P=0.65)

Total (95% CI) 0.00 [-0.51, 0.51]
Test for heterogeneity: [I.sub.2]=28.6%
Test for overall effect: Z=0.00 (P=1.00)

WMD=weighted-mean-differences, CI=confidence interval.

Figure 5: Proportion of patients that
required allogeneic red blood cell transfusion.

Outcome: Proportion transfused

Study Gelatin Control Weight
 n/N n/N %

01 Gelatin vs Crystalloid
Innerhofer 2002 (18) 2/20 2/20 2.13
Parker 2004 (26) 31/198 22/198 21.95
Mittermayr 2007 (22) 8/21 1/20 0.75
Soares 2009 (28) 2/20 2/20 2.13
Schramko 2010 (9) 4/15 3/15 2.60
Subtotal (95% CI) 274 273 29.56
Total events: 47 (Gelatin), 30 (Control)
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=2.05 (P=0.04)

02 Gelatin vs Albumin
Niemi 2006 (8) 3/15 5/15 4.73
Subtotal (95% CI) 15 15 4.73
Total events: 3 (Gelatin), 5 (Control)
Test for heterogeneity: not applicable
Test for overall effect: Z=0.82 (P=0.41)

03 Gelatin vs Starch
Innerhofer 2002 (18) 2/20 1/20 1.06
Van der Linden 2004 (34) 2/55 9/55 10.26
Van der Linden 2005 (35) 21/68 24/64 20.22
Niemi 2006 (8) 3/15 10/15 9.46
Mittermayr 2007 (22) 8/21 3/19 2.31
Ooi Su Min 2009 (25) 42/45 40/45 3.15
Vanhoonacker 2009 (36) 15/72 20/82 17.51
Schramko 2010 (9) 4/15 2/15 1.73
Subtotal (95% CI) 311 315 65.71
Total events: 97 (Gelatin), 109 (Control)
Test for heterogeneity: [I.sup.2]=50.8%
Test for overall effect: Z=1.17 (P=0.24)

Total (95% CI) 8/22 603 100.00
Total events: 147 (Gelatin), 144 (Control)
Test for heterogeneity: [I.sup.2]=42.3%
Test for overall effect: Z=0.24 (P=0.81)

Study OR (fixed)
 95% CI

01 Gelatin vs Crystalloid
Innerhofer 2002 (18) 1.00 [0.13, 7.89]
Parker 2004 (26) 1.49 [0.83, 2.67]
Mittermayr 2007 (22) 11.69 [1.30, 105.03]
Soares 2009 (28) 1.00 [0.13, 7.89]
Schramko 2010 (9) 1.45 [0.26, 8.01]
Subtotal (95% CI) 1.67 [1.02, 2.73]
Total events: 47 (Gelatin), 30 (Control)
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=2.05 (P=0.04)

02 Gelatin vs Albumin
Niemi 2006 (8) 0.50 [0.10, 2.63]
Subtotal (95% CI) 0.50 [0.10, 2.63]
Total events: 3 (Gelatin), 5 (Control)
Test for heterogeneity: not applicable
Test for overall effect: Z=0.82 (P=0.41)

03 Gelatin vs Starch
Innerhofer 2002 (18) 2.11 [0.18, 25.35]
Van der Linden 2004 (34) 0.19 [0.04, 0.94]
Van der Linden 2005 (35) 0.74 [0.36, 1.53]
Niemi 2006 (8) 0.13 [0.02, 0.66]
Mittermayr 2007 (22) 3.28 [0.72, 14.94]
Ooi Su Min 2009 (25) 1.75 [0.39, 7.81]
Vanhoonacker 2009 (36) 0.82 [0.38, 1.74]
Schramko 2010 (9) 2.36 [0.36, 15.45]
Subtotal (95% CI) 0.79 [0.53, 1.17]
Total events: 97 (Gelatin), 109 (Control)
Test for heterogeneity: [I.sup.2]=50.8%
Test for overall effect: Z=1.17 (P=0.24)

Total (95% CI) 1.04 [0.77, 1.40]
Total events: 147 (Gelatin), 144 (Control)
Test for heterogeneity: [I.sup.2]=42.3%
Test for overall effect: Z=0.24 (P=0.81)

OR=odds ratio, CI=confidence interval.

Figure 6: Total blood loss.

Outcome: Total Blood Loss

 Gelatin
Study N Mean (SD) N

01 Gelatin vs Crystalloid
Tollofsrud 1995 (31) 10 1398.00 (918.00) 10
Wahba 1996 (37) 10 620.00 (256.00) 10
Innerhofer 2002 (18) 20 611.00 (270.00) 20
Liang 2006 (21) 20 409.00 (124.00) 20
Mittermayr 2007 (22) 21 526.00 (388.00) 20
Soares 2009 (28) 20 1000.00 (489.00) 20
Jin 2010 (19) 12 314.00 (58.00) 12
Schramko 2010 (9) 15 1099.00 (420.00) 15
Subtotal (95% CI) 128 127
Test for heterogeneity: [I.sup.2]=14.3%
Test for overall effect: Z=1.27 (P=0.20)

02 Gelatin vs Albumin
Tollofsrud 1995 (31) 10 1398.00 (918.00) 10
Wahba 1996 (38) 10 620.00 (256.00) 10
Karoutsos 1999 (20) 15 1064.00 (385.00) 12
Niemi 2006 (8) 15 1070.00 (422.50) 15
Trekova 2006 (32) 18 894.00 (310.00) 18
Subtotal (95% CI) 68 65
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=1.41 (P=0.16)
03 Gelatin vs Starch
Mortelmans 1995 (24) 21 2778.00 (956.00) 21
Wahba 1996 (38) 10 620.00 (256.00) 10
Karoutsos 1999 (20) 15 1064.00 (385.00) 15
Innerhofer 2002 (18) 20 611.00 (270.00) 20
Van der Linden 2004 (34) 55 661.00 (304.00) 55
Van der Linden 2005 (35) 68 1498.00 (1131.00) 64
Liang 2006 (21) 20 409.00 (124.00) 20
Niemi 2006 (8) 15 1070.00 (423.00) 15
Boks 2007 (15) 90 1921.00 (844.00) 90
Mittermayr 2007 (22) 21 526.00 (388.00) 19
Yap 2007 (40) 20 561.00 (227.00) 20
Ooi Su Min 2009 (25) 45 596.00 (337.00) 45
Vanhoonacker 2009 (36) 72 592.00 (348.00) 82
Jin 2010 (19) 12 314.00 (58.00) 12
Schramko 2010 (9) 15 1099.00 (420.00) 15
Subtotal (95% CI) 499 503
Test for heterogeneity: [Chi.sup.2]=19.58, df=14 (P=0.14), I==28.5%
Test for overall effect: Z=0.71 (P=0.48)

04 Gelatin vs Dextran
Tollofsrud 1995 10 1398.00 (918.00) 10
Subtotal (95% CI) 10 10
Test for heterogeneity: not applicable
Test for overall effect: Z=0.10 (P=0.92)

Total (95% CI) 705 705
Test for heterogeneity: [I.sup.2]=18.8%
Test for overall effect: Z=0.56 (P=0.57)

 Control Weight
Study Mean (SD) %

01 Gelatin vs Crystalloid
Tollofsrud 1995 (31) 1131.00 (677.00) 0.14
Wahba 1996 (37) 473.00 (278.00) 1.26
Innerhofer 2002 (18) 577.00 (228.00) 2.89
Liang 2006 (21) 375.00 (128.00) 11.35
Mittermayr 2007 (22) 296.00 (254.00) 1.73
Soares 2009 (28) 1058.00 (518.00) 0.71
Jin 2010 (19) 321.00 (84.00) 20.75
Schramko 2010 (9) 921.00 (367.00) 0.87
Subtotal (95% CI) 39.69
Test for heterogeneity: [I.sup.2]=14.3%
Test for overall effect: Z=1.27 (P=0.20)

02 Gelatin vs Albumin
Tollofsrud 1995 (31) 1080.00 (467.00) 0.17
Wahba 1996 (38) 668.00 (398.00) 0.8
Karoutsos 1999 (20) 905.00 (411.00) 0.75
Niemi 2006 (8) 840.00 (335.00) 0.93
Trekova 2006 (32) 865.00 (280.00) 1.86
Subtotal (95% CI) 4.51
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=1.41 (P=0.16)
03 Gelatin vs Starch
Mortelmans 1995 (24) 3437.00 (1578.00) 0.11
Wahba 1996 (38) 668.00 (398.00) 0.8
Karoutsos 1999 (20) 821.00 (374.00) 0.94
Innerhofer 2002 (18) 656.00 (269.00) 2.48
Van der Linden 2004 (34) 814.00 (577.00) 2.33
Van der Linden 2005 (35) 1533.00 (972.00) 0.54
Liang 2006 (21) 412.00 (105.00) 13.65
Niemi 2006 (8) 1140.00 (305.00) 0.99
Boks 2007 (15) 1768.00 (712.00) 1.33
Mittermayr 2007 (22) 319.00 (435.00) 1.05
Yap 2007 (40) 507.00 (183.00) 4.24
Ooi Su Min 2009 (25) 567.00 (281.00) 4.21
Vanhoonacker 2009 (36) 696.00 (369.00) 5.39
Jin 2010 (19) 349.00 (98.00) 16.67
Schramko 2010 (9) 951.00 (336.00) 0.93
Subtotal (95% CI) 55.67
Test for heterogeneity: [Chi.sup.2]=19.58, df=14 (P=0.14), I==28.5%
Test for overall effect: Z=0.71 (P=0.48)

04 Gelatin vs Dextran
Tollofsrud 1995 1436.00 (793.00) 0.12
Subtotal (95% CI) 0.12
Test for heterogeneity: not applicable
Test for overall effect: Z=0.10 (P=0.92)

Total (95% CI) 100
Test for heterogeneity: [I.sup.2]=18.8%
Test for overall effect: Z=0.56 (P=0.57)

 WMD (fixed)
Study 95% CI

01 Gelatin vs Crystalloid
Tollofsrud 1995 (31) 267.00 [-439.96, 973.96]
Wahba 1996 (37) 147.00 [-87.23, 381.23]
Innerhofer 2002 (18) 34.00 [-120.88, 188.88]
Liang 2006 (21) 34.00 [-44.10, 112.10]
Mittermayr 2007 (22) 230.00 [30.17, 429.83]
Soares 2009 (28) -58.00 [-370.20, 254.20]
Jin 2010 (19) -7.00 [-64.76, 50.76]
Schramko 2010 (9) 178.00 [-104.26, 460.26]
Subtotal (95% CI) 27.04 [-14.72, 68.79]
Test for heterogeneity: [I.sup.2]=14.3%
Test for overall effect: Z=1.27 (P=0.20)

02 Gelatin vs Albumin
Tollofsrud 1995 (31) 318.00 [-320.36, 956.36]
Wahba 1996 (38) -48.00 [-341.30, 245.30]
Karoutsos 1999 (20) 159.00 [-144.37, 462.37]
Niemi 2006 (8) 230.00 [-42.87, 502.87]
Trekova 2006 (32) 29.00 [-163.98, 221.98]
Subtotal (95% CI) 89.19 [-34.63, 213.01]
Test for heterogeneity: [I.sup.2]=0%
Test for overall effect: Z=1.41 (P=0.16)
03 Gelatin vs Starch
Mortelmans 1995 (24) -659.00 [-1448.10, 130.10]
Wahba 1996 (38) -48.00 [-341.30, 245.30]
Karoutsos 1999 (20) 243.00 [-28.63, 514.63]
Innerhofer 2002 (18) -45.00 [-212.04, 122.04]
Van der Linden 2004 (34) -153.00 [-325.36, 19.36]
Van der Linden 2005 (35) -35.00 [-394.13, 324.13]
Liang 2006 (21) -3.00 [-74.21, 68.21]
Niemi 2006 (8) -70.00 [-333.91, 193.91]
Boks 2007 (15) 153.00 [-75.13, 381.13]
Mittermayr 2007 (22) 207.00 [-49.51, 463.51]
Yap 2007 (40) 54.00 [-73.79, 181.79]
Ooi Su Min 2009 (25) 29.00 [-99.20, 157.20]
Vanhoonacker 2009 (36) -104.00 [-217.31, 9.31]
Jin 2010 (19) -35.00 [-99.43, 29.43]
Schramko 2010 (9) 148.00 [-124.19, 420.19]
Subtotal (95% CI) -12.84 [-48.10, 22.42]
Test for heterogeneity: [Chi.sup.2]=19.58, df=14 (P=0.14), I==28.5%
Test for overall effect: Z=0.71 (P=0.48)

04 Gelatin vs Dextran
Tollofsrud 1995 -38.00 [-789.86, 713.86]
Subtotal (95% CI) -38.00 [-789.86, 713.86]
Test for heterogeneity: not applicable
Test for overall effect: Z=0.10 (P=0.92)

Total (95% CI) 7.56 [-18.75, 33.87]
Test for heterogeneity: [I.sup.2]=18.8%
Test for overall effect: Z=0.56 (P=0.57)

WMD=weighted-mean-differences, CI=confidence interval.
COPYRIGHT 2012 Australian Society of Anaesthetists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2012 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Saw, M.M.; Chandler, B.; Ho, K.M.
Publication:Anaesthesia and Intensive Care
Article Type:Report
Geographic Code:1USA
Date:Jan 1, 2012
Words:11382
Previous Article:Risks of anaesthesia and surgery in elderly patients.
Next Article:The role of regional anaesthesia techniques in the management of acute pain.
Topics:

Terms of use | Privacy policy | Copyright © 2021 Farlex, Inc. | Feedback | For webmasters