Efficacy of high voltage pulsed current for healing of pressure ulcers in patients with spinal cord injury.

Pressure ulcers are a potentially serious complication in patients with spinal cord injury (SCI). Such ulcers are a main cause of lengthened hospitalization, delay of rehabilitation, and hospital readmission for patients with SCI. (1-3) Although theoretically preventable, pressure ulcers develop in 50% to 60% of patients with complete SCI. (4) Mawson et al (5) reported that 59% of patients admitted to the hospital following SCI developed a pressure sore within 30 days, and Richardson and Mayer (4) recorded a 41% incidence of pressure sores in patients with SCI at the time of admission to a rehabilitation facility.

     Table 1. Grades of Pressure Ulcer Severity (a)
  Grade I: An acute inflammatory response involving the epidermis; the epidermis
remains intact.  Healing may occur in several days.
  Grade II: A break in or blistering of the epidermis, surrounded by erythema
and induration.  Healing may require up to 2 weeks.
  Grade III: A shallow, irregular defect extending through the dermis to the
subcutaneous fat junction.  Healing may require months.
  Grade IV: The ulcer extends through the full thickness of the skin into the
subcutaneous tissue, fascia, or muscle.  Undermining may be present.  Healing
may require months or surgery.
  Grade V: Penetration of the ulcer extensively into the underlying bone, with
no anatomic limit.  The ulcer has exposed bone, joint, muscle, or
fascia at its base.  Healing usually requires surgery.

Pressure ulcers are notoriously slow-healing in patients with SCI, (6) and medical management is expensive. Nursing costs increase 50% when pressure ulcers develop. (7) The projected cost per year of treatment of early-occurring pressure ulcers in patients with SCI is approximately $66 million. (5) The economic effects of pressure ulcers extend beyond direct wound care, however, because the patient is prevented from pursuing gainful employment or participating fully in rehabilitation activities. Methods to speed healing of pressure ulcers in patients with SCI are thus of interest to the patients and their families, physicians and rehabilitation team members, third-party payers, and society.

A pressure ulcer is defined as "an ulceration of skin and/or deeper tissues due to unrelieved pressure, shear force(s), and/or frictional force(s)." (8)(p63) Significant etiologic factors in pressure ulcer occurrence after SCI include paralysis and sensory loss. (2,9) Other factors contributing to pressure ulcer occurrence are soft tissue atrophy, increased age, poor diet, smoking, anemia, vitamin deficiency, hypoproteinemia, and skin maceration attributable to incontinence. (2,9-13) Bacterial contamination, although not a direct cause of pressure ulcers, can cause breakdown and prolonged healing of soft tissue. (8)

Pressure ulcers have been classified into stages, according to severity of tissue damage. (7) Expected healing rate is partially related to severity of tissue damage and varies from weeks to months (Tab. 1). Conservative (non-surgical) management of established pressure sores involves control of the causal factors, such as removal of pressure, avoidance of skin maceration, correction of nutritional deficiencies, removal of necrotic tissue, control of infection, and encouragement of soft tissue repair. (2,3,7,12) A plethora of methods for accelarating granulation and reepithelialization of chronic wounds exist in medical practice. Electrical stimulation is one method advocated for facilitation of soft tissue healing.

Electrical Stimulation in

Animal and In Vitro Studies

Many controlled studies of soft tissue healing using animal models have indicated that electrical stimulation applied as either low intensity direct current (LIDC) (14-24) or high voltage pulsed direct current (HVPC) (25-28) is effective. Young (28) reported that application of HVPC following a 12-hour period of tourniquet-induced ischemia prevented the gangrene noted in the legs of control dogs. In animal models, HVPC has been found relatively ineffective in accelerating healing of incisional skin wounds, (25,26) but effective in promoting tendon repair. (27) Brown and colleagues (25,26) noted a tendency for beneficial effects to occur when negative polarity was used the first 4 postoperative days, followed by positive polarity for days 4 to 7 after skin wounding. Owoeye et al (27) reported that using positive polarity for the HVPC wound electrode produced significantly stronger tendon repair than did the use of negative polarity.

In in vitro studies, stimulation of biosynthesis using both LIDC (29,30) and HVPC (31) has been reported. Bactericidal effects of LIDC (32,33) and HVPC (34) have also been noted in in vitro studies. Efficacy of electrical stimulation for cellular biosynthesis and bacterial control may have optimal thresholds for intensity, treatment time duration, frequency, and polarity. Bourguignon and Bourguignon (31) found a maximal effect on DNA and protein synthesis in cultured human fibroblasts using an HVPC intensity of 50 to 75 V, a stimulator frequency of 100 pps, and a negative electrode polarity. Maximal bactericidal effects were found using HVPC with an intensity of 250 V at the cathode for a treatment period of 2 hours. (34) Guffey and Asmussen (33) found that a 30-minute application of HVPC produced no bactericidal effect at any intensity and hypothesized that a treatment time greater than 30 minutes may be required to produce a bactericidal effect in vitro with HVPC.

Results of animal and in vitro studies may not be completely applicable to wound-healing problems encountered in clinical practice. In all of the animal experiments we reviewed, with the exception of that of Young, (28) who induced injury by circulatory constriction, the wound studied was created by sharp surgical incision. Furthermore, all of the animals were healthy, except in the study of Smith et al, (22) who studied healing of skin incisions in mice with induced diabetes. In clinical practice, problems with soft tissue healing frequently occur, not in healthy skin that has been recently surgically incised, but in chronic craterlike skin defects of patients having vascular or neurological impairment. Unfortunately, soft tissue healing in humans cannot be studied under the same controlled circumstances as in studies involving animal models.

Electrical Stimulation

Efficacy for Healing in

Human Subjects

Beneficial effects of healing of chronic ulcers in human subjects have been reported for both LIDC (35-40) and HVPC (41-45) stimulation. Thurman and Christian (44) reported healing of an infected plantar ulcer in a diabetic patient after 4 months of HVPC treatment. In two other clinical reports, HVPC was reported successful for healing plantar ulcers in a diabetic population (42) and for healing chronic pressure ulcers in a geriatric population. (45) Kloth and Feedar, (43) in a controlled, randomized study of chronic grade IV ulcers in patients having no neurological impairment, reported complete healing of all ulcers receiving HVPC, whereas control group ulcers increased in size. Akers and Gabrielson (41) studied the healing rate of ulcers in patients having either SCI or denervation and found patients receiving HVPC had a faster healing rate than patients receiving either whirlpool or whirlpool combined with HVPC.

Treatment protocols for application of electrical stimulation to human wounds have varied widely. In all of the studies cited previously, an electrode was placed directly over the ulcer. Some evidence exists that electrical stimulation using alternate electrode placement sites, such as over the base of a skin flap, (46) on the hand, (47) or epidurally, (48) may stimulate a healing response. Studies comparing various electrode sites for efficacy of human skin ulcer healing have not appeared in the literature. Protocols have included initiating HVPC treatment with positive (43) or negative (45) electrode polarity and changing polarity after specified time intervals, as well as maintaining positive polarity until healing occurred. (42) Treatment time duration for HVPC stimulation was reported as 20 minutes a day, (44) 45 minutes a day, (43) and 30 minutes twice a day. (45) Intensity and frequency settings reported for HVPC stimulation have been between 100 to 175 V and 50 to 105 pps, respectively, with no muscle contraction produced. (42,43,45) No studies have yet been conducted, to our knowledge, that identify maximally effective HVPC intensity, frequency, polarity, and treatment time characteristics for healing chronic ulcers of specific etiologies in human subjects.

Electrical stimulation does appear to have a potentially beneficial effect on healing of chronic ulcers in humans, based on the literature reviewed. The majority of publications on the subject, however, are in the form of clinical reports, with few controlled, randomized studies containing a statistical analysis of the results. The ulcers included in each report have been of mixed etiologies and varied locations, and in none of these studies was the study sample exclusively representative of patients with SCI. Some anecdotal evidence indicates patients with SCI or sensory loss respond less well to LIDC than do patients without such neurological impairment. (39,40) Akers and Gabrielson, (41) however, reported that the patients receiving HVPC, who demonstrated the highest healing rate, had either SCI or partial denervation. Considering the reported efficacy of electrical stimulation in accelerating healing of chronic ulcers and the significant medical costs of treatment of pressure ulcers, further controlled, clinical study of the efficacy of HVPC in healing pressure ulcers in patients with SCI seemed warranted.

The objective of this study was to assess the efficacy of HVPC for healing of pressure ulcers in patients with SCI. The null hypothesis was that the average percentage of decrease from pretreatment ulcer size in the HVPC group would be equal to that in the placebo group after 5, 10, 15, and 20 days of treatment. The alternative hypothesis was that the HVPC group would demonstrate a greater percentage of decrease from pretreatment ulcer size than the placebo group.

Method and Materials

Subjects

The study sample consisted of inpatients from the Spinal Cord Injury Service at Baptist Memorial Hospital Regional Rehabilitation Center (Memphis, Tenn). Patients were invited to participate in the study at the time of their admission to the unit if they met the following criteria: The patient had to be male, have a diagnosis of complete or incomplete SCI, and have a pelvic pressure ulcer (over either the sacral/coccygeal or gluteal/ischial regions) classified between grades II and IV (DeLisa Classification System, Tab. 1). Only male patients were included in this study, because hospital records indicated that predominately male patients were admitted to this service. We anticipated a small sample size and wanted to control for as many variables as possible. Patients were excluded if they had severe cardiac disease, cardiac arrhythmia, or uncontrolled autonomic dysreflexia or if they used a pacemaker. Participants (patient or legal guardian) signed an informed consent form after a complete explanation of the purpose and procedures of the study. There was no charge to patients for participation in the study, and no renumeration was given. A total of 20 patients met the inclusion criteria and were selected to participate in the study.

Procedure

On entry to the study, patients were stratified into groups according to ulcer classification (grade II, III, or IV) and smoking status (smoker or nonsmoker), because these two factors are known to affect the rate of expected healing. (7,10) If a patient had multiple pelvic ulcers, the largest ulcer in terms of wound surface area (WSA) was selected for study. Patients were then randomly assigned to either an HVPC group or a placebo HVPC group.

Patients received the respective treatment for 1 hour a day, for 20 consecutive days. We set the treatment time at 1 hour a day, so that treatment could be reasonably fit into the patients' busy rehabilitation schedule. We set the study end point at 20 days because we wanted to compare the effect on healing time for the two groups after a specific time interval and because we estimated the minimal stay for patients would be 20 days. All treatments were administered by one of three persons--two physical therapists (JWG and JKC) and a nursing coordinator (RAM). The nursing staff and patients were kept blinded as to patient treatment group assignment. We used this modified double-blind design as a control for the known placebo effect on healing of pressure ulcers. (49) Patients were dropped from the study if they developed medical complications requiring transfer from the rehabilitation center to the hospital or if surgical closure of the ulcer was performed. Measurements of WSA were conducted before the first treatment and after the 5th, 10th, 15th, and 20th treatment days.

Treatment Protocols

Electrical stimulation was applied using an Intelect 500 HVPC stimulator. (*) The stimulator produces a twin-peaked pulse, with approximately 75-[mu]s spacing between pulses, and has an adjustable output of 0 to 500 V and an adjustable frequency of 1 to 120 pps. The ulcer was covered with gauze soaked with sterile 0.9% saline; deep ulcers were packed with saline-soaked gauze. A piece of heavy-duty aluminum foil, cut slightly larger than the ulcer perimeter, was attached with an alligator clip to the negative lead of the HVPC unit. The foil electrode was placed over the ulcer on top of the saline-soaked gauze, as described by Unger. (45) The saline-soaked gauze was then covered with a piece of plastic wrap to retard drying, followed by a dry gauze pad, which was taped to the skin. A sandbag was used, if needed, to hold the wound electrode in place. The dispersive electrode (20X25 cm) was strapped over the patient's medial thigh, with a wet cloth placed between the electrode and the patient's skin. The stimulator frequency was set at 100 pps, with a continuous mode. The intensity was slowly increased to 200 V. The ammeter on the HVPC unit was checked to ensure current was flowing, and contact between the patient's skin and the electrodes was adjusted, if necessary, to obtain maximum current flow for the 200-V intensity used. This intensity of stimulation applied did not evoke obvious muscle contractions in any patient, and no patient complained of any discomfort during treatment. The patient remained in the same position for the 60-minute treatment.

The placebo treatment procedure was the same as that described for the HVPC treatments. The electrode lead, however, was taped outside of the alligator-clip receptacle so that no current flowed, although the HVPC unit appeared to be working, because the light, voltmeter, and timer were functioning. Similar taping was used around the junction of the alligator clips for both HVPC and placebo group patients.

The rationale for our electrical stimulation protocol was as follows. We set the ulcer electrode at negative polarity in reference to the dispersive electrode, partially because other investigators (36,37,45) recommended initiating treatment with negative polarity, particularly when infection is suspected. (36,37,45) We anticipated that the pretreatment wound-culture study would indicate some bacterial presence in most cases. (50) Although observations regarding polarity effects in LIDC stimulation may not apply to HVPC stimulation, (33) our impression from the reviewed literature was that negative polarity was more often effective than positive polarity. Although some investigators (40,43,45) recommended changing polarity after 3 to 7 days or when healing plateaus, the rationale for this change was not clear to us, nor was an operational definition of "healing plateau" clear. We therefore maintained the wound electrode at negative polarity for the entire 20 days of study, to investigate the effect on healing when polarity of the wound electrode remains negative for a specific time period. Some evidence exists that patients with denervation may require a higher intensity of stimulation than other patients for maximal healing responses, (36,41) so we sought to produce a total current of approximately 500 [microamp], based on the reports of Cheng et al (30) and Carley and Wainapel. (36) To achieve this output on the high voltage stimulator used in this study, a pulse rate of 100 pps at 200 V is required. (51) We placed the dispersive electrode over the medial thigh, primarily because of convenience. Many of the patients had corsets or braces on the torso, so placing the dispersive (anode) electrode proximal to the wound electrode, as recommended by Kloth and Feedar, (43) was not always feasible. We therefore selected a dispersive electrode site that would be identical for all patients.

All patients received equivalent nursing care. Cleansing of ulcers was performed twice a day by the nursing staff using Cara-Klenz, (+) followed by an application of Carrington gel (+) and a dry dressing. Wounds were mechanically debrided, as necessary; enzymatic debridement was not used. All ulcers were cultured before treatment began. All possible efforts were made to keep pressure off the ulcer. A routine 2-hour turning schedule was followed when patients were in bed. Patients with sacral/coccygeal ulcers were not allowed to lie supine. Patients with gluteal/ischial ulcers were not allowed to sit until their ulcers were well-healed, with the exception of patient 7, who was allowed to sit an average of 9 hours a day during the study period. The type of bed mattress and wheelchair cushion was not changed for any patient during the course of the 20-day study.

Nutritional status of the patients was evaluated by a nutritionist and a physician upon admission to the unit. Nutritional status and dietary intake were monitored on a regular basis. The regular diet planned for all patients was sufficient to meet caloric, protein, and vitamin requirements. Patients found to have anemia or hypoproteinemia were given a diet or a dietary supplement appropriate to correct the condition.

Measurement Procedures

Measurements of WSA were conducted before the first treatment and after every fifth treatment for 20 days. At each measurement session, three 35-mm color slides were taken at a distance of 27.9 to 30.5 cm (11-12 in), with a metric ruler taped next to the ulcer. The camera used was an Olympus OM-2s (+2) with a 50-mm 3.5 macro lens.

To obtain measurements of WSA, each slide was projected onto paper and the focus adjusted so that the metric ruler measurements on the slide corresponded with a superimposed ruler identical to the one in the slide. The ulcer perimeter was carefully traced and then transferred to a tablet digitizer with a stylus pen[section]; the area (in square millimeters) was calculated using the Generic CADD 3.0 software program (II) interfaced with an IBM-PC/XT. (#) The WSA values for the three slides from each measurement session were averaged, giving for each patient one average WSA measurement for days 0, 5, 10, 15, and 20. The same person (JWG) conducted all WSA measurements.

This method of WSA measurement, using digitizer analysis of slides, has been described previously. (41,42) To verify the reliability of measurements obtained with this method, we performed a test-retest study of six pressure ulcers before the study began. Three slides were taken of each ulcer, and the procedure was repeated after an hour. Slides were analyzed as described earlier. An analysis of test-retest measurements was conducted using an intraclass correlation coefficient (ICC[3,1]), (52) with an obtained value of .99. To further verify consistency of the procedure of tracing slides and digitizing, midway through the study, the analysis of 11 patients' slides was repeated on a subsequent day. Comparing measurements between the two days, an ICC (3,1) value of .99 was obtained. We therefore considered the WSA measurements in this study to be reliable.

At each measurement session, a Tracing of ulcer perimeter onto a clear transparency was made using a permanent felt-tip pen. The tracings were used for immediate feedback to patients concerning ulcer size. The feedback was conducted in a standardized way to avoid bias.

Results of laboratory blood studies, wound cultures, and other patient information were obtained from the patients' medical records. Wound culture results were classified as either "positive" or "negative." Anemia was defined as having a hematocrit of less than 36%, and hypoproteinemia was defined as having a serum albumin concentration of less than 3.0 g/dL.

Data Reduction and Analysis

For the day-5, -10, -15, and -20 measurements on each patient, absolute change (in square millimeters) from pretreatment ulcer size was calculated. Because pretreatment WSA differed among patients, measurements at days 5, 10, 15, and 20 were calculated as the percentage of change from the pretreatment (baseline) WSA.

The Mann-Whitney U Test for two independent samples was used to compare the percentage of change from the pretreatment WSA of the two groups at days 5, 10, 15, and 20 (one-sided) test). We used a nonparametric test to avoid assuming sizes of ulcers are normally distributed or have equal variances at all times and for both treatments. We performed separate tests for each day rather than a repeated-measures analysis of variance (ANOVA), because each measure was dependent on previous measures. For example, if an ulcer had healed by day 15, the outcome for day 20 was also known. This specific relationship among the repeated measures is not considered in the usual model for a repeated-measures ANOVA and would violate the assumption of independence among observations.

Our reasons for using a one-sided test were as follows. Our objective a priori was to learn whether ulcer healing rate was accelerated by adding HVPC to good nursing care. Appropriate care would be expected to decrease the size of pressure ulcers. Our null hypothesis was that the average decrease in ulcer size in the group given HVPC would be equal to the average decrease in ulcer size in the placebo group. The appropriate alternative hypothesis was that the average decrease in ulcer size would be greater with HVPC than without HVPC. The question of whether HVPC might actually have the opposite effect (ie, slower healing) was not considered because (1) past studies have provided evidence that one might expect faster healing, (2) there is no evidence from past studies that would suggest electrical stimulation delays or slows healing, and (3) that alternative was inconsistent with our research objective and was of no interest to us in this study. Prior to the study, sample size calculations indicated a sample size of 10 in each group was sufficient for 80% power to detect a difference of 20% improvement between groups using a one-sided test, given a standard deviation of 15%.

To assess the significance of differences between groups in continuous variables, the Mann-Whitney U Test was used. To assess the significance of differences between groups in characteristics

[TABULAR DATA OMITTED]

having a nominal scale, the Fisher's Exact Test was used. A significance level of .05 was used for all statistical tests.

Results

Seventeen of the 20 patients completed the study. Three patients were transferred to the acute care hospital (2 patients developed medical complications, and 1 patient required surgical repair of his ulcer) and thus were eliminated from the study. No significant differences in characteristics between the HVPC and placebo groups existed, except that the HVPC group demonstrated a significantly longer duration of SCI than did the placebo group (Tab. 2).

Percentage-of-change decreases in WSA exhibited by the HVPC group were significantly greater than in the placebo group at day 5 (P = .03), day 15 (P = .05), and day 20 (P = .05). Differences at day 10 did not reach significance (P = .14). The null hypothesis was thus rejected for the day-5, -15, and -20 measurements but accepted for the day-10 measurements.

The healing rates of individual patients over the 20 days of treatment, expressed as percentages of change from pretreatment (day 0) WSA, are presented in the Figure. Grade II ulcers in both groups healed completely. All grade III ulcers in the HVPC group showed consistent patterns of decreased WSA over the treatment period; two grade III ulcers in the placebo group demonstrated periods of increased WSA, whereas the other three grade III ulcers in that group demonstrated consistent decreases WSA. The grade IV ulcer receiving HVPC showed a 67% decrease in WSA by the 20th day of treatment. The grade IV ulcer receiving placebo treatment showed little change over the treatment period, with only a 15% decrease from its original size at the 20th day. The median percentage of change from the initial WSA and the minimum and maximum amounts of change in each group at each measurement interval are presented in Table 3. At the end of the 20 days of study, the worst response in the HVPC group was a 52% decrease; the worst response in the placebo group was a 14% decrease from pretreatment WSA.

Discussion

In this study, we investigated the efficacy of a specific protocol of HVPC for healing of pelvic pressure ulcers in patients with SCI. Our results indicated that the mean percentage of decrease in WSA in the HVPC group was significantly greater than that of the placebo group at the 5th, 15th, and 20th days of treatment.

Results of this study concerning the efficacy of HVPC for promoting healing of chronic ulcers in humans confirm the observations of other researchers. [41-45] Our findings are somewhat difficult to compare with the findings of the two other controlled clinical trials of HVPC efficacy for chronic ulcers. Akers and Gabrielson [41] did not indicate the location of pressure sores studied or the diagnoses of patients studied, although they did indicate that patients in the HVPC group had sensory loss. Kloth and Feedar [43] studied dermal ulcers

Table 3. Median, Minimum, and Maximum Percentage of Change from Pretreatment (Day 0) Wound Surface Area in High Voltage Pulsed Direct Current (HVPC) and Placebo Groups at Each 5-Day Measurement Point

                       Group
Measurement Time       HVPC       Placebo
Day 5
  Median % change       -32        -14
  Minimum % change      -12        +17
  Maximum % change     -100        -74
Day 10
  Median % change       -47        -42
  Minimum % change        -23        +42
  Maximum % change        -100        -41
Day 15
  Minimum % change        -42         +22
  Maximum % change     -100       -100
Day 20
  Median % change          -80        -52
  Minimum % change         -52        -14
  Maximum % change        -100       -100

representative of a wide variety of locations and etiologies in patients with intact peripheral nerves. Our study is the only controlled study of HVPC to our knowledge in which the type of ulcer studied has been limited to pelvic pressure ulcers in patients with SCI. Our study protocol cannot be compared with that of Akers and Gabrielson because no description of that protocol was given. Kloth and Feedar used a similar frequency (105 pps), a somewhat lower intensity (100-175 V), and a different polarity and location of the dispersive electrode as compared with our study. They began treatment with the wound electrode at positive polarity, changed polarity to negative when ulcer size plateaued, and alternated polarity daily when another healing plateau occurred. The authors also recommended placing the anode cephalad and closer to the neuraxis than the cathode. We kept the wound electrode polarity negative for the 20 days of study and placed the dispersive (positive) electrode distal to the ulcer (over the medial thigh). Kloth and Feedar treated ulcers 5 days a week until healing occurred, whereas we treated ulcers for 20 consecutive days. All of the users in Kloth and Feedar's study were grade IV ulcers, whereas only two ulcers in our study were that severe. They reported a mean length of treatment of 7.3 weeks and a mean healing rate of 45% a week. We found a mean percentage of decrease of 80% from the original ulcer size in the HVPC group after 20 days of treatment.

All patients in the HVPC group demonstrated consistent decreases in ulcer size over the 20 treatment days, although thore were individual differences in rate of decrease. We did not observe any periods of plateau in rate of decrease of ulcer size in the patients in the HVPC group. Conversely, some ulcers of the patients in the placebo group demonstrated plateaus as well as increases in WSA. Changes in WSA of the HVPC-treated ulcers might have been different if we had alternated positive and negative polarities daily or placed the dispersive electrode in a different location. It is also possible, however, that polarity and location of the dispersive electrode make no difference in the healing response. Nonetheless, our data represent the first documentation of the rate of WSA change using HVPC with negative wound electrode polarity at specific intensity and frequency settings over a 20-day period. Further clinical research is needed to compare the efficacy of various HVPC protocols, in terms of polarity, intensity, and frequency, for promoting ulcer healing in specific patient populations.

Although the specific means by which electrical stimulation may promote healing of soft tissue are not known, many hypotheses exist such as attraction of connective tissue and inflammatory cells, modification of endogenous electrical potentials of tissue, stimulation of cellular biosynthesis and replication, bactericidal effects, enhanced circulation, and generation of a cellular electrophysiological effect. [19,31,42] Electrochemical effects at electrode sites, such as increased cell temperature, changes in pH, or release of ions from electrodes, may be possible during LIDC stimulation, but these reactions were found not to be produced during in vitro studies with HVPC. [31,33,34] Newton and Karselis [53] reported no change in skin pH of healthy subjects after HVPC treatment with cathodal current. Because we did not attempt to investigate the mechanism(s) by which HVPC might be effective, it is not possible for us to speculate about the mechanism(s) responsible for the improvement that we found. Future studies are needed to investigate the mechanism(s) involved.

A possible limitation of this study is that the placebo group had the greatest number of sacral/cocygeal ulcers and the HVPC group had the most gluteal/ischial ulcers. The most common location for pelvic ulcers in patients with SCI is unclear from the literature. Both sacral [4,5] and ischial [54,55] sites have been identified as the most frequent, although the incidence has also been reported as equal at sacral and ischial sites. [2] We made the assumption in planning the study that ischial and sacral ulcers would occur equally and that, through random assignment, ulcer locations would be equally distributed in the two treatment groups. That assumption proved untrue, probably because of the small sample size. A possible alternative explanation for our results is that gluteal/ischial ulcers heal faster than do sacral/coccygeal ulcers; however, we do not believe that to be the case. The healing rate of the three HVPC group sacral/coccygeal ulcers (patients 2, 3, 8; Figure) did not appear to be different from that of the gluteal/ischial ulcers. Indeed, patient 3 had a sacral grade II ulcer that was completely healed by treatment day 10. Few data exist in the literature to document the expected healing rate of ischial versus sacral pressure ulcers in patients with SCI. Apple and Murray [56] stated that ischial ulcers heal faster than trochanteric ulcers, but no mention was made of sacral healing rate, nor was statistical evidence of healing presented. Noble [55] reported only a slightly longer mean treatment time for sacral (79 days) than ischial (74 days) ulcers.

Another possible limitation to our study was the chance inclusion of both patients older than 70 years in the placebo group (patients 10 and 15). Patient 10, however, had a WSA decrease of 43% by day 10, indicating that appropriate healing had begun with good nursing care. Patient 15 had a healing profile almost identical to that of 26-year-old patient 9. Therefore, age, by itself, did not appear to have a confounding effect.

The placebo effect on ulcer healing is improvement attributable to patient and nursing staff enthusiasm about participating in an experiment concerning a novel healing agent. [49] We tried to control for this effect by blinding of nursing staff and patients as to treatment. We expected that the placebo effect would be operating in all patients and that the effect of HVPC would be added to the placebo effect in the HVPC group. We believe the significant effects on healing were due to a combination of HVPC, placebo effect, and excellent nursing care. The changes in WSA exhibited by the ulcers of the placebo group in this study may be greater than those demonstrated by similar ulcers of patients receiving nursing care outside of an "experimental" environment.

In planning this study, we stratified patients by ulcer severity and by patient smoking status, because these were the two factors from the literature known to effect ulcer healing rate. We feel that future studies would be strengthened by also stratifying patients according to location of ulcer before random assignment to treatment groups. The design of future studies might also be improved by having the person conducting the measurements blinded as to ulcer treatment. This procedure was not possible in this study because of personnel limitations and the length of the study (2.5 years). Further research is needed to explore the relative efficacy of various electrical stimulation protocols, so that maximally effective treatment protocols for particular patient condition can be identified.

The clinical implications of this study are that HVPC may increase that healing rate of pelvic pressure ulcers in patients with SCI and that HVPC may be a significant adjunct to good nursing care in promoting healing of such ulcers. Prevention of pressure ulcers remains the goal of medical care for patients with SCI. If pressure ulcers do occur, however, prompt treatment to stimulate rapid healing is indicated. The addition of HVPC treatment to a wound care program may promote more rapid healing, thus decreasing treatment time, treatment costs, and length of institutional stay of patients with SCI who develop pressure ulcers.

Conclusion

In this study of male patients with pelvic pressure ulcers, ulcers treated with HVPC demonstrated a significantly greater percentage of decrease from their original size after the 5th, 15th, and 20th days of treatment than did placebo ulcers. The clinical implications of this study are that HVPC can be a significant adjunct to good medical care in promoting healing of pelvic ulcers in patients with SCI.

References

[1] LEvinson W, Ward G, Valleroy M. Care of spinal-cord-injured patients after the acute period. Spine. 1989;3:495-516.

[2] Constable JD, Pierce DS. Pressure sores. In: Pierce DS, Wickel VH, eds. The Total Care of Spinal Cord Injuries. Boston, Mass: Little, Brown & Co Inc; 1977:187-204.

[3] Kahn S. A guide to the treatment of decubitus (pressure) ulcers in paraplegia. Surg Clin North Am. 1960;40:1657-1675.

[4] Richardson RR, Mayer PR. Prevalence and incidence of pressure sores in acute spinal cord injuries. Paraplegia. 1981;19:235-247.

[5] Mawson AR, Biundo JJ, Neville P, et al. Risk factors for early-occurring pressure ulcers following spinal cord injury. Am J Phys Med Rehabil. 1988;67:123-127.

[6] Basson MD, Burney RE. Defective wound healing in patients with paraplegia and quadriplegia. Surg Gynecol Obstet. 1982;155:9-12.

[7] DeLisa JA, Mikulic MA. Pressure ulcers: what to do if preventive management fails. Postgrad Med. 1985;77:209-220.

[8] Crenshaw RP, Vistnes LN. A decade of pressure sore research: 1977-1987. J Rehabil Res Dev. 1989;26:63-74.

[9] Daniel RK, Priest DC, Wheatley DC. Etiologic factors in pressure sores: an experimental model. Arch Phys Med Rehabil. 1981;62:492--497.

[10] Lamid S, Ghatit AZ. Smoking, spasticity and pressure sores in spinal cord injury patients. Am J Phys Med Rehabil. 1983;62:300-306.

[11] Maklebust J. Pressure ulcers: etiology and prevention. Nurs Clin North Am. 1987;22:359--377.

[12] Skylar CG. Pressure ulcer management in the neurologically impaired patient. J Neurosci Nurs. 1985;17:30-36.

[13] Thiyagarajan C, Silver JR. Aetiology of pressure sores in patients with spinal cord injury. Br Med J [Clin Res]. 1984;289:1487-1490.

[14 Akai M, Oda H, Shirasaki Y, Tateishi T. Electrical stimulation of ligament healing: an experimental study of the patellar ligament of rabbits. Clin Orthop. 1988;235:296-301.

[15] Assimocopoulos D. Wound healing promotion by the use of negative electrical current. Am Surg. 1968;34:423-431.

[16] Alvarez OM, Merta PM, Smerbeck RV, Eaglestein WH. The healing of superficial skin wounds is stimulated by external electric current. J Invest Dermatol. 1983;81:144-148.

[17] Bigelow JB, Al-Husseini SA, VonRecum AF, Park JB. Effect of electrical stimulation of canine skin and percutaneous device-skin interface healing. In: Brighton CT, Black J, Pollack S eds. Electrical Properties of Bone and Cartilage. New York, NY: Gruce & Stratton Inc; 1979:289-310.

[18] Carey LC, Lepley D. Effect of continuous direct current on healing wounds. Surgical Forum. 1962;13:33-35.

[19] Dunn MG, Doillon CJ, Berg RA, et al. Wound healing using a collagen matrix: effect of direct current electrical stimulation. J Biomed Mater Res. 1988;22:191-206.

[20] Konikoff JJ. Electrical promotion of soft tissue repairs. Ann Biomed Eng. 1976;4:1-5.

[21] Rowley BA, McKenna JM, Chase JR. The influence of electrical current on an infecting microorganism in wounds. Ann NY Acad Sci. 1974;238:543-551.

[22] Smith J, Romansky N, Vomero J, Davis RH. The effect of electrical stimulation on wound healing in diabetic mice. J Am Podiatr Assoc. 1984;74:71-75.

[23] Stanish WD, Rubinovich M, Kozey J, MacGillvary G. The use of electricity in ligament and tendon repair. The Physician and Sportsmedicine. 1985;13:109-116.

[24] Wu KT, Go N, Dennis C, et al. Effect of electric currents and interfacial potentials on wound healing. J Surg Res. 1967;7:122-128.

[25] Brown M, Gogia PP. Effects of high voltage stimulation on cutaneous wound healing in rabbits. Phys Ther. 1987;67:662-667.

[26] Brown M, McDonnell MK, Menton DN. Electrical stimulation effects on cutaneous wound healing in rabbits: a follow-up study. Phys Ther. 1988;68:955-960.

[27] Owoeye I, Spielholz NI, Nelson AJ. Low-intensity pulsed galvanic current and the healing of tenotomized rat Achilles tendons: preliminary report using load-to-breaking measurements. Arch Phys Med Rehabil. 1987;68:415-418.

[28] Young HG. Electric impulse therapy aids wound healing. Modern Veterinary Practice. 1966;47:60-62.

[29] Nessler JP, Mass DP. Direct-current electrical stimulation of tendon healing in vitro. Clin Orthop. 1987;287:303-312.

[30] Cheng N, Hoof VH, Bockx E, et alk. The effects of electric currents on ATP generation, protein synthesis and membrane transport in rat skin. Cli Orthop. 1982;171:264-272.

[31] Bourguignon BJ, Bourguignon LY. Electric stimulation of protein and DNA synthesis in human fibroblasts. FASEB J. 1987;1:398-902.

[32] Barranco SD, Spadero JA, Berger JJ, Becker RO. In vitro effect of weak direct current on Sta Staphylococcus aureus. Clin Orthop. 1974; 100:250-255.

[33 Guffey JS, Asmussen MD. In vitro bactericidal effects of high voltage pulsed current versus direct current against Staphylococcus aureus. Clinical Electrophysiology. 1989;1:5-9.

[34] Kincaid CB, Lavoie KH. Inhibition of bacterial growth in vitro following stimulation with high voltage, monophasic, pulsed current. Phys Ther. 1989;69:651-655.

[35] Assimacopoulos D. Low intensity negative electrical current in the treatment of ulcers of the leg due to chronic venous insufficiency. Am J Surg. 1969;115:683-687.

[36] Carley PJ, Wainapel SF. Electrotherapy for acceleration of wound healing: low intensity direct current. Arch Phys Med Rehabil. 1985;66:443-446.

[37] Gault WR, Gatens PF Jr. Use of low intensity direct current in management of ischemic skin ulcers. Phys Ther. 1976;56:265-269.

[38] Mulder GD, LaParr M. Decubitus ulcer: update on new approaches to treatment. Geriatrics. 1988;43:37-50.

[39] Wheeler PC, Wolcott LE, Morris JL, Spangler MR. Neural considerations in the healing of ulcerated tissue by clinical electrotherapeutic application of weak direct current: findings and theory. In: Reynolds DV, Sjoberg AE, eds. Neuroelectric Research. Springfield, Ill: Charles C Thomas, Publisher; 1971:83-89.

[40] Wolcott LE, Wheeler PC, Hardwick HM, Rowley BA. Accelerated healing of skin ulcers by electrotherapy: preliminary clinical results. South Med J. 1969;52:795-801.

[41] Akers TK, Gabrielson AL. The effect of high voltage galvanic stimulation on the rate of healing of decubitus ulcers. Biomed Sci Instrum. 1984;20:99-100.

[42] Alon G, Stein H. Diabetic ulcer healing using high voltage TENS. Phys. Ther. 1986; 66:775. Abstract.

[43] Kloth LC, Feedar JA. Acceleration of wound healing with high voltage, monophastic, pulsed current. Phys Ther. 1988;88:503-508.

[44] Thurman BF, Christian EL. Response of a serious circulatory lesion to electrical stimulation. Phys Ther. 1971;51:1107-1110.

[45] Unger PG. Wound healing using high voltager galvanic stimulation. Stimulus. 1985;10:8-10.

[46] Lundeberg T, Kjartansson J, Samuelson U. Effect of electrical nerve stimulation on healing of ischeamic skin flaps. Lancet. 1988;2:712-714.

[47] Kaada B. Promoted healing of chronic ulceration by transcutaneous nerve stimulation (TNS). Vasa. 1983;12:262-269.

[48] Augustinnson LE, Carlsson CA, Holm J, Jivegard L. Epidural electrical stimulation in severe limb ischemia. Ann Surg. 1985;202:104-110.

[49] Fernie GR, Dornan J. The problems of clinical trials with new systems for preventing or healing decubiti. In: Denedi RM, Cowden JM, eds. Bedsore Biomechanics. Baltimore, MD: University Park Press; 1975;315-320.

[50] Paris LC, Witkowski JA, Crissey JT, eds. The Decubitus Ulcer. New York, NY: Masson Publishing USA Inc; 1983:31.

[51] Alon G, DeDomenico G. High Voltage Stimulation an Integrated Approach to Clinical Electrotherapy. Chattanooga, Tenn: Chattanooga Corp; 1987.

[52] Shrout PE, Fleiss JL. Intraclass correlations: uses in assessing rater reliability. Psychol Bull. 1979;86:420-428.

[53] Newton RA, Karselis TC. Skin pH following high voltage pulsed galvanic stimulation. Phys Ther. 1983;63:1593-1596.

[54] Constantian MB, Jackson HD. The ischial ulcer. In: Constantian MB, ed. Pressure Ulcers: Principles and Techniques of Management. Boston, Mass: Little, Brown & Co Inc; 1980:215.

[55] Noble PC. The Prevention of Pressure Sores in Persons with Spinal Cord Injuries. New York, NY: World Rehabilitation Fund Inc; 1981.

[56] Apple DF, Murray HH. Pressure sores: factors influencing care, Paraplegia. 1984;22:51-52.

(*) Chattanooga Corp, PO Box 4287, Chattanooga, TN 37405.

(1) Carrington Laboratories Inc, Dallas, TX 75356-9500.

(2) Olympus Optical Co, Tokyo, Japan.

(3) Hitachi Seiko Ltd, Tokyo 101, Japan.

(4) Generic Software Inc, Redmond, WA 98052.

(5) International Business Machines Corp, Old Orchard Rd, Armonk, NY 10504.

JW Griffin, MS, PT, is Associate Professor, Department of Rehabilitation Sciences, College of Allied Health Sciences, The Health Sciences Center, The University of Tennessee, Memphis, 822 Beale St, Memphis, TN 38163 (USA). Address all correspondence to Ms Griffin.

RE Tooms, MD, is Professor, Department of Orthopaedic Surgery, The University of Tennessee, Memphis.

RA Mendius, MS, CRRN, is Rehabilitation Coordinator, Baptist Memorial Hospital Regional Rehabilitation Center, 1025 E H Crump Blvd, Memphis, TN 38104.

JK Clifft, MS, PT, is Associate Professor, Department of Rehabilitation Sciences, The University of Tennessee, Memphis.

R Vander Zwaag, PhD, is Associate Director of Health Services Research, Baptist Memorial Hospital, 899 Madison Ave, Memphis, TN 38146.

F El-Zeky, PhD, is Research Associate, Health Service Research, Baptist Memorial Hospital.

This study was aproved by the Baptist Memorial Hospital Patient Participation Committee.

This study was funded in part by a grant from the Foundation for Physical Therapy Inc.