Acute abdomen; pre and post-laparotomy diagnosis.Introduction
The acute abdomen may be defined generally as an intra abdominal process causing severe pain and often requiring surgical intervention. It is a condition that requires a fairly immediate judgment or decision as to management (Module 2, 2008:1).
Abdominal pain is a common presentation to emergency department. It is vital that the physician has an understanding and be familiar with the presentations of common diseases that cause abdominal pain (Laurell H, 2006:2, Flasar MH, 2006:3).
Preoperative diagnosis of acute abdomen is crucial to minimize the morbidity and mortality especially where the diagnostic facilities are limited (Chhetri RK, 2005:4).
While most of the etiologies of acute abdomen are not life threatening, rapid diagnosis and therapy may be life saving in some cases. Preoperative accurate diagnosis prevents from unnecessary laparotomies and results in reducing negative operations (Saleh M Abbas, 2007:5).
Previous studies have shown that a considerable volume of diagnostic errors would be reduced by paying more attention to diagnosis before laparotomy (Gauderer MW, 1997:7).
Abdominal Pain represents 5% of emergency room visits. Only 10% of these evaluations require surgery. Acute abdomen accounts for 10% of malpractice claims (Graff, 2001:6).
Nowadays, despite availability of different diagnostic tools and progress in new imaging methods like ultrasonography(US) and computed tomography scanning, correct pre-operative diagnosis of acute abdomen still remains challenging.
The diagnosis of acute abdomen is not always straightforward and an accurate diagnostic approach is required to get the right decision.
Improvement in the surgeons' power of decision making in confrontation with such patients is the basic pivot of disease diagnosis and therapy, particularly in developing countries with limited diagnostic facilities (Chhetri RK, 2005:4).
A few studies considering the accuracy of pre-operative diagnosis has been performed. The goal of this study is to compare pre and post laparotomy diagnosis and to identify the rate of negative laparotomies as to guide practicing surgeons confronted with acute abdomen.
Materials and Methods
This was an observational study performed in emergency surgical ward of Sina hospital (Tehran University of Medical Sciences; Tehran, Iran) from February to December 2005, to compare the pre-operative diagnosis based on clinical examination and evaluations with the post-operative diagnosis of acute abdomen.
The study included 139 cases of all age groups and both genders with clinical manifestations suggestive of acute abdomen that underwent laparotomy. The excluded patients were those who had a history of trauma (traumatic acute abdomen). Case series method was considered as the method of sampling.
Patients were examined by the admitting surgical team after taking a thorough history, Relevant points in the history included the patient's gender, site of pain, character of pain, fever, loss of appetite, change in bowel habit, vomiting, abdominal distension and urinary or genital symptoms. Factors in the clinical examination that were considered of significant contribution to the final diagnosis included temperature, tachycardia, and abdominal tenderness and localized or generalized guarding.
In all studied cases, white blood cell (WBC) count with a differential leukocyte count (DLC) and measurement of neutrophil percent were performed on admission. Urinalysis (UA) performed for 95% of patients. Abdomen X-ray, US and serum amylase level measurements were performed in some cases considering the clinical suspicion. Pre-operative diagnosis was made by surgical residents based on clinical examination and investigations compared to the post operative diagnosis.
Rate of negative laparotomy, sensitivity, specificity, positive and negative predictive values considering leukocytosis (WBC count [greater than or equal to] 11,000 per micro liter in peripheral blood smear), granulocytosis (neutrophils >75% in DLC), UA (considered positive if contained [greater than or equal to] 5 WBC or [greater than or equal to] RBC or showed pregnancy), US and X-ray were all calculated. Statistical analysis was performed using SPSS software version 11.5. Student's t-test and Chi-square test were used to calculate the significance level and a P-value of <0.05 was considered significant.
Total 139 patients diagnosed with acute abdomen underwent emergency laparotomy. Ninety (64.7%) were male and 49(35.3%) were female. Mean age of the patients was 35.3 [+ or -] 18.6 with the range of 9-85 years. Sixty-eight patients (49%) were 20-29 years old.
The most common symptoms in our patients with abdominal pain were nausea (69.1%) and vomiting (43.9%). The most common clinical signs were abdominal tenderness (97.1%), voluntary guarding (66.9%) and rebound tenderness (66.2%). The signs and symptoms of the patients are summarized on table 1.
Acute appendicitis was the most common cause of acute abdomen (56.8%).Acute appendicitis was the etiology of acute abdomen in 67% of male and 38.8% of female patients. Other common causes of acute abdomen were peritonitis (14.4%) and bowel obstruction (7.9%) in male, and torsion of an ovarian cyst (24.5%) in female patients. The pre and post laparotomy diagnosis are reported in table 2.
Leukocytosis and granulocytosis were observed in 66.2% and 80% of patients, respectively. Eighty percent of patients suffering from peritonitis and 77.5% of appendicitis patients had leukocytosis. Granulocytosis had the highest sensitivity (79.3%).
UA was positive in 20.1% of patients. One had ectopic pregnancy. Urinalysis had the highest negative predictive value (91%).
Abdominal X-ray was requested for 54 (38.9%) patients. In 10 patients (18.5%) abnormal findings were present. Abdominal X-ray was performed for 100% of patients with bowel obstruction and 80% of them were found to be positive. Overall X-ray had the highest specificity (88.8%) and the lowest sensitivity (46.6%) and negative predictive value (25%). US was performed in 72(51.8%) patients. They were 41(56.9%) female and 31(43.1%) male patients. Fifty-eight (80.6%) patients had positive findings. Overall US performed for 83.7% of female and 34.4% of male patients. Correct diagnosis in 100% of patients with cholecystitis and ovarian torsion and 68.6% of patients with appendicitis, was performed with US. US had the highest positive predictive value (97.6%). Serum amylase level was measured in 62 patients (44.6%). Liver function tests were requested for 27.7 of patients. Sensitivity, specificity, positive and negative predictive value of leukocytosis, granulocytosis, urinalysis, and abdominal X-ray, US and serum amylase level are summarized in table 3.
Total negative laparotomy rate was 12.2% (P value < 0.05). Comparison of pre and post laparotomy diagnoses is shown in table 4.
In 77.7% of patients, pre and post laparotomy diagnosis were the same. The diagnostic accuracy rates were 92.2% and 79.6% in male and female patients, respectively. All of the patients with rupture of abdominal aortic aneurysm had correct pre-operative diagnosis. In 88.8% of patients with appendicitis and 87.5% of patients with cholecystitis, both the pre and post-operative diagnoses were the same. Pre-operative diagnosis was correct in only 50% of ovarian cyst torsion.
Despite improvement in clinical evaluations and advancement in diagnostic methods, correct diagnosis of acute abdomen is still sometimes difficult. Patients with acute abdominal pain are a heterogeneous group that consumes a great deal of a surgical department's resources (Saleh M Abbas, 2007:5). In cases when the diagnosis is suspected, laparotomy has been advised to be performed (Scott Hs, 1993:8), but this policy has increased the rate of negative laparotomies (Tadvrel P, 1992:9).
In this study, acute abdomen was most common in 20-29 years (49% of patients). This result is similar to statistics from other studies, reporting the prevalence of acute abdomen mostly in 20-29 years old patients (Chhetri RK, 2005:4). The causes of acute abdomen are several and their relative incidence varies in different populations. Several factors are described to be responsible for these differences. Socioeconomic factors and diet have mostly been incriminated to be responsible for the observed differences (Kotiso, 2006:10).
Among the etiologies leading to laparotomy, in this study acute appendicitis was the commonest and observed in 56.8% of cases. Peritonitis and bowel obstruction were observed in 14.4% and 7.9% of cases respectively. Other studies, reported acute appendicitis to be the leading cause of acute abdomen in 55% cases (Chhetri RK, 2005:4), visceral perforation and bowel obstruction in 8-12% and 15-24% of cases of laparotomy, respectively (Heelar M, 1997:11).
The most frequently ordered study for abdominal pain is the CBC. The CBC should never be used to make the sole diagnosis; however, because nearly 11% of normal adults have an elevated WBC count and 13% have left shifts (Bohrn M, 2004:12). In our study, the sensitivity and specificity of leukocytosis were 70% and 84.5% respectively and of granulocytosis were 79.3% and 83.3% executively.
Other studies reported the sensitivity of leukocytosis equal to 77-87% and the specificity equal to 63-67%. Sensitivity and specificity of granulocytosis in other reports were 91.5% and 64.5%, respectively (Chhetri RK, 2005:4). In our study, none of these tests had the required sensitivity and specificity to predict of acute abdomen etiology.
Urinalysis was performed for 95% of patients and in 28 (20.1%) cases had positive findings. In a study performed, urinalysis had sensitivity and specificity 75% and 84% respectively (Chhetri RK, 2005:4) and in our study, 78% and 81% respectively. Regarding previous studies, UA is advised to be performed for all acute abdomen patients to exclude urinary tract infection (UTI), diabetes, renal stones, ectopic pregnancy and normal pregnancy (Heelar M, 1997:11).
Plain abdominal radiography performed for 54 patients (41.5%), it had the most accuracy of diagnosis in mechanical bowel obstruction with sensitivity of 83.3% and specificity of 97%. X-ray had the highest specificity (88.8%) and the lowest sensitivity (46.6%). Chhetri reported sensitivity of 64.8% and specificity 88.8% for plain abdominal X-ray (Chhetri RK, 2005:4). Bowel obstruction is usually confirmed by abdominal radiography in decubitus (horizontal) and upright positions. In these positions most of the findings are as follow: intestinal loop caliber >3 centimeters, air-liquid level and gas increase in colon.
In our study, US was performed for 72 patients (51.8%).The sensitivity and specificity of US was 79% and 73% respectively. In our study the diagnostic accuracy for cholecystitis was as high as 100%. Chhetri reported sensitivity and specificity of 69.4% and 81.5 for US in the diagnosis of acute abdomen and the diagnostic accuracy of 95% for cholecystitis (Chhetri RK, 2005:4).
Serum amylase measurement was performed in 62 patients (44.6%). Its sensitivity and specificity was 74% and 50% respectively.
In our study negative appendectomy rate was 13.2% which is similar to the statistics presented by other studies between 15- 30% (Boleslawski E, 1999:13, John PF, 1990:14). The negative laparotomy rate was 12.2% in this study. Overall accuracy rate was 69.8%. In this group of patients, the pre and post-laparotomy diagnosis were the same. In other investigations, the overall accuracy of diagnosis has been reported 80% by skilled physicians and 50% by young physicians (Paterson-brown S, 1991:15). In the study of Chhetri, negative laparotomy rate was 17.6 % (Chhetri RK, 2005:4).
Acute abdomen diagnosis is based on complete history taking, physical examination and investigation tools including laboratory tests and radiological findings. The investigative modalities are good guidance and helpful to confirm the diagnosis. For example, when suspicious to intestinal obstruction, one can perform abdominal X-ray which would be a great help in diagnosis confirmation or sonographic guidance for the diagnosis of cholecystitis. High levels of serum amylase may guide our suspicion toward pancreatitis. A preoperative accurate diagnosis prevents from negative laparotomies.
Authors would like to thank the head of Sina Trauma and Surgery Research Center, Professor Moosa Zargar for his guidance to perform this study. This research has been funded by the Sina trauma and research center of Tehran University of Medical Sciences. Authors would like to confirm that there has been no conflict of interests in this research.
Bohrn M, Siewert B (2004), "Acute abdominal pain: What not to miss", Patient Care, Vol. 38, pp. 31-39.
Boleslawski E, Panis Y, Benoisr S et al (1999), "Plain abdominal radiography as a routine procedure for acute abdominal pain of lower quadrant: prospective evaluation", World J Surg, Vol. 13 No. 3, pp. 262-264.
Chhetri RK, Shrestha ML (2005), "A Comparative study of preoperative with operative diagnosis in acute abdomen", Kathmanolu university Med. J, Vol. 3 No. 2, pp. 107-110.
Flasar MH, Goldberg E. (2006), "Acute abdominal pain", Med Clin North Am, Vol. 90, pp.481-503.
Gauderer MW. (1997),"Acute abdomen-when to operate immediately and when to observe", Semin Pediatr surg, Vol. 6 No. 2, pp.74-80
Graff (2001), "Acute abdominal pain", Emerg. Med Clin North Am, Vol.19, pp.123-136 John, PF (1990), "Practicality in the management of the acute abdomen", Br J Surg, Vol. 77 No. 41, pp. 364-7.
Kotiso, Abdurahman Z (2006), "Pattern of Acute Abdomen in Adult Patients in Tikur Anbessa Teaching Hospital, Addis Ababa, Ethiopia", East and Central African J Surg, Vol. 12 No. 1, pp.47-52.
Laurell H, Hansson LE, Gunnarsson U. (2006), "Diagnostic pitfalls and accuracy of diagnosis in acute abdominal pain", Scand J Gastroenterol., Vol. 41, pp. 1126-31.
MODULE 2(2008) -Diseases and Malfunctions. MINICOURSE 4-THE ACUTE ABDOMEN. http://www.ece.ncsu.edu/ imaging/MedImg/SIMS/Module2/GE2_4.html. (Assessed on: 31.1.2008).
Paterson-brown S (1991), "Strategies for reducing inappropriate laparotomy rate in the acute abdomen", BMJ, Vol. 303 No. 6810, pp.1115-8.
Saleh M Abbas, Smithers T., Truter E (2007), "What clinical and laboratory parameters determine significant intra abdominal pathology for patients assessed in hospital with acute abdominal pain?", World J Emerg. Surg., Vol. 2 No. 26, pp. 1749-7922.
Scott Hs, Rosin Rd (1993), "The influence of diagnostic and therapeutic laparoscopy on patients presenting with acute abdomen", J. R. Soc Med, Vol. 86 No. 12, pp. 699-701.
Tadvrel P, Born MP, Pradel J et al (1992), "Acute abdomen of unknown origin: impact of CT on diagnosis and management", Gastrointestinal Radiol, Vol. 17 No. 4, pp.287-91.
Tarraza, H.M. Robert D, Moor (1997), "Gynaecologic causes of the acute abdomen and the acute abdomen in pregnancy", Surg clin of North Am, Vol.77, No.6, pp.1371-1394.
Marjan Laal (MD, Specialist of general surgery)
Sina Trauma and Surgery Research Center
Department of Surgery, Faculty of Medicine
Tehran University of Medical Sciences (TUMS), Iran
Armita Mardanloo (MD)
Sina Trauma and Surgery Research Center
Tehran University of Medical Sciences (TUMS), Iran
Corresponding author: email@example.com
Correspondence concerning this article should be addressed to Dr. Marjan Laal, Sina Trauma and Surgery Research Center, Department of Surgery, Faculty of Medicine, Tehran University of Medical Sciences (TUMS), Iran Postal Box: 11365-3876
Table 1: Signs and symptoms of patients presented with acute abdominal feature Symptom N (%) Abdominal pain 139 (100%) Severity of pain: Mild 22 (15.8%) Moderate 93 (66.9%) Severe 24 (17.3%) Character of pain: Colic 53 (38.1%) Continuous 86 (61.9%) Nausea 96 (69.1%) Vomiting 61 (43.9%) Loss of appetite 59 (42.4%) Bowel habit(+) 16 (11.5%) Abdominal 14 (10.1%) distention Sign N (%) Abdominal 135 (97.1%) tenderness Rebound 92 (66.2%) tenderness Voluntary 93 (66.9%) guarding Generalized 27 (19.4%) guarding Localized 84 (60.4%) guarding Pulse Rate [greater than or equal to] 110 50 (36%) Temperature [greater than or equal to] 38 30 (21.6%) Table 2: The causes of acute abdominal pain Pre-operative N (%) diagnosis Appendicitis 84 (60.4%) PeriStonitis 12 (8.6%) Cholecystitis 9 (6.5%) Ovarian cyst torsion 5 (3.6%) Ruptured * AAA 5 (3.6%) Intestinal obstruction 3 (2.2%) Other disease 30 (15.1%) Post-operative diagnosis N (%) Appendicitis 79 (56.8%) Peritonitis due to: 20 (14.4%) Perforated appendicitis 5 (3.5%) Perforated peptic ulcer 8 (5.7%) Pancreatitis 2 (1.4%) Perforated cholecystitis 1 (0.7%) Infected uterine carcinoma 1 (0.7%) ** AMI 3 (2.2%) Ovarian cyst torsion 12 (8.6%) Cholecystitis 10 (7.2%) Intestinal obstruction due 10 (7.2%) to: Adhesion 3 (2.2%) Volvolus 3 (2.2%) Incarcerated hernia 2 (1.4%) Tumor 1 (0.7%) Invagination 1 (0.7%) Ruptured AAA. 4 (2.9%) Ectopic pregnancy 1 (0.7%) Peritoneal hematoma 1 (0.7%) Aortodeodenal fistula 1 (0.7%) Abdominal wall abscess 1 (0.7%) * AAA = Abdominal aortic aneurysm ** AMI = acute mesenteric ischemia Table 3: Predictive values of investigations Leukocytosis granulocytosis Sensitivity (%) 70 79.3 Specificity (%) 84.5 83.3 [??] / PPV (%) 96 97 # NPV (%) 80 60 * UA ** US X-Ray Amylase Sensitivity (%) 78 79 46.4 74 Specificity (%) 81 73 88.8 50 [??] / PPV (%) 68 97.6 95.4 71 # NPV (%) 91 60 25 54 * UA = Urinalysis ** US = Ultrasonography, [??] / PPV = Positive predictive value # NPV = Negative predictive value Table 4: Comparison between pre and post-operative diagnosis Final diagnosis Exact ** PO One of PO None of PO Peritonitis 9 (45%) 11 (55%) 0 (0%) Appendicitis 70 (88.6%) 9 (11.4%) 0 (0%) Cholecystitis 8 (80%) 1 (10%) 1 (10%) Ovarian cyst torsion 2 (16.7%) 7 (58.3%) 3 (25%) Intestinal 4 (40%) 4 (40%) 2 (20%) obstruction Rupture of * AAA 4 (100%) 0 (0%) 0 (0%) * AAA = Abdominal aortic aneurysm ** PO = Pre-operative diagnosis