British Journal of Plastic Surgery (1997), 50, 37k-379 0 1997 The British Association of Plastic Surgeons
Ischial pressure sore coverage: a rationale for flap selection
R. D. Foster, J. P. Anthony, S. J. Mathes and W. Y. Hoffman
Division of Plastic and Reconstructive Surgery, University of California at San Francisco, California, USA
SUMMARY The role of wound debridement and flap coverage in treating pressure sores is clearly established. However, criteria and supportive clinical data for specific flap selection and the sequence of flaps for coverage of the ischium remain ill-defined. From 1979-1995, 114 consecutive patients underwent flap coverage of 139 ischial pressure sores. Preoperative risk factors, prior flap history, defect size, flap success, complication rates, and the length of hospitalization were retrospectively evaluated and compared for 112 flaps in 87 patients. Flap success was defined as a completely healed wound. Average follow-up was 10 months (range: 1 month-9 years).
Overall, 83% (93/l 12) of the flaps healed. In the majority of cases (75%, 84/l 12), wound debridement and flap reconstruction was achieved in a single stage. However, there were significant differences in the healing rates among the various flaps used. The inferior gluteus maximus island flap and the inferior gluteal thigh flap had the highest success rates, 94% (32/34) and 93% (25/27), respectively, while the V-Y hamstring flap and the tensor fascia lata flap had the poorest healing rates, 58% (7/12) and 50% (6/12), respectively. Flap success was not significantly affected by the age of the patient or the prior number of flaps used and preoperative risk factors were equally distributed across all types of flaps. The overall complication rate was 37% (41/112), most commonly from a slight wound edge dehiscence (n = 16) that healed with local wound care within one month postoperatively.
Results of this study show that proper flap selection and the appropriate sequence of flap use significantly improve success rates for ischial pressure sore coverage in both the short- and long-term. Based upon flap reliability (successful healing rates), reusability, and the need to preserve as many future flap options as possible, a rationale for flap selection is presented which can be individualized to any patient.
Accepted current surgical management of pressure sores consists of thorough debridement including involved bone, followed by flap coverage. Despite this and other non-surgical measures, pressure sore recurrence rates can still exceed 50%.* Frustration with these results has prompted a subset of surgeons to suggest that perhaps some or all of these patients have little to gain from any operative treatment.* However, much of the data to support this pessimistic view is based on initial operative success rates of only 80%.* Since pressure sores which fail to heal while the patient is hospitalized virtually never heal after discharge, the early postoperative results will clearly contribute to future outcomes.
Our experience has led us to believe that the most important step for early treatment success is appropri- ate flap selection. The spectrum of flaps used for ischial coverage includes the gluteus maximus, poster- ior thigh,4 hamstring,5 biceps fernoTis tensor fascia lata, gracilis, rectus abdominis and anterior thigh flaps.O Despite this wide variety of coverage options, the ischium remains the most difficult pressure sore site to treat.
The optimal approach to pressure sore coverage should include utilization of the most reliable flap to achieve a healed wound, while preserving as many future flap options as possible. Despite the widespread use of myocutaneous flaps in the treatment of pressure sores, few studies have critically analysed flap reliability, comparing one flap to another. The purpose of this study was to examine our 17-year surgical experience
with flap reconstruction for ischial pressure sore coverage and to compare and contrast flap reliability rates in obtaining a healed wound. Treatment efficacy was also analysed from several other perspectives including a flaps versatility for reuse, thereby main- taining as many future flap options as possible. Based upon these results, we have substantiated a rationale for flap coverage designed to treat a patient over his/her lifetime.
Patients and methods
From 1979 to 1995, 114 consecutive patients under- went flap coverage of 139 ischial pressure sores. From this group, there was adequate information available to permit detailed analysis of patients records for 112 flaps in 87 unselected patients (64 men, 23 women). Several patients had bilateral and/or recurrent ischial sores. The mean age of the patients was 49 years (range 16-90 years).
Chronic wounds (present for > 3 months) repre- sented 52% (58/112) of the total number of cases. In most cases the patients were paraplegic (89%) or quadraplegic (4%) and only 7% were ambulatory. In 75% (84/l 12) of the cases, the patients had their pres- sure sore debrided and reconstructed in a single stage. Patients were confined to bedrest postoperatively, usu- ally on an air-fluid bed, for lo-14 days followed by a programme of gradually increased weight-bearing (sitting) on the ischial area.
Ischial pressure sores 375
Table 1 Flaps selected and primary healing rates
Flap No. of eases Healed primarily (S) (54)
Inferior gluteus maximus island 34 (30.4) 94 Gluteal thigh 21(24.1) 93 Gracilis 16 (14.3) 75 V-Y hamstring 12 (10.7) 58 Tensor fascia lata 12 (10.7) 50 Anterior thigh 6 (5.3) 100 Rectus abdominis 5 (4.5) 100
Flap success and primary healing were defined as a healed wound within one month postoperatively, usually corresponding to the time of wound suture removal during the patients first follow-up visit. Reconstructive failure was defined as a case resulting in a non-healed wound. Follow-up ranged from 1 month to 9 years (average 10.7 months).
The flap selection for ischial sore coverage is listed in Table 1. Only musculocutaneous and/or fasciocuta- neous flaps were included in this series. The inferior gluteus maximus island flap was most commonly used (30.4%, 34/l 12) followed by the inferior gluteal thigh flap (24.1%, 27/112). Five of the six ambulatory patients were reconstructed with gracilis flaps. Flap success (healing) was compared between each flap. Risk factors for impaired wound healing, based on prior wound healing studies,2.z and complication rates were determined for each reconstruction. The age of the patient, number of prior flaps and defect size following debridement were analysed independently to test their effect on subsequent flap success. Significance was determined post hoc by the Chi squared test.
Overall, 83% (93/l 12) of the flaps healed primarily. 89% (75/84) of the cases treated in a single stage (debridement and flap reconstruction) healed primar- ily. For grossly infected (necrotic tissue) sores treated in a single stage, flap success was also 89%.
For flaps in which we had a significant experience (>lO cases), the inferior gluteus maximus island flap and the inferior gluteal thigh flap, had the highest success rates, 94% (32/34) and 93% (25/27), respectively (Table l), followed by the gracilis flap, the V-Y ham- string flap and the tensor fascia lata flap. The hospital stay averaged 21 days, 16.5 days for patients without prior complicated medical histories. The time to heal averaged 38 days. The earliest time healing could have occurred, as defined in this study, was 28 days (the earliest time of suture removal).
Complications occurred in 41 (37%) of the recon- structions (Table 2). There were no intraoperative deaths or significant donor site complications. A slight wound edge dehiscence (not > 1 cm in length or width) occurred in 16 patients but all of these wounds healed with local wound care alone within one month post- operatively. When flaps failed to heal, requiring further operative management, it was usually the result of
Table 2 Complications
Slight wound edge dehiscence (healed with local wound care) 16 Partial flap necrosis 10 Wound infection 5 Wound dehiscence (requiring reoperation) 5 Died, unrelated causes 2 Aspiration pneumonia 1 Intraop. myocardial infarct 1 Deep venous thrombosis 1
either partial flap necrosis (n = lo), inadequate debridement and the subsequent failure to adequately control infection (n = 5) or wound dehiscence (n = 5). Reoperation following instances of treatment failure involved either readvancement of a flap or harvesting of a different flap. The results of these decisions are included in the flap selection totals in Table 1. Flap necrosis was seen exclusively in one of three flaps: the tensor fascia lata flap (n = 5), gracilis (n = 3), and the V-Y hamstring (n = 2). Flap necrosis of the tensor fascia lata and gracilis flaps was always confined to the distal one third of the flap. Three out of five cases of wound dehiscence that ultimately failed to heal involved the proximal suture line across the V-Y ham- string flap. The other two cases of dehiscence were attributed to excessive tension with the wound closure. None of the ambulatory patients experienced signili- cant limitations to their range of motion or gait in the long term.
Among individual flaps, the inferior gluteus maxi- mus island flap appears to be very reliable. Although 6 cases resulted in a slight wound dehiscence success- fully managed with local wound care, no specific fac- tor(s) could be identified that would limit this flaps use. The only failed flap was the result of inadequate debridement. The inferior gluteal thigh flap was also very reliable. The two cases of failed reconstructions with the inferior gluteal thigh flap were the result of inadequate debridement (1 patient) and extremely poor patient compliance (1 patient) resulting in wound breakdown. A slight wound edge dehiscence did not prevent healing in four additional patients.
In 67 cases (60%) prior flap reconstruction had been performed. Overall, 84% (38/45) of the flaps in patients without any prior reconstruction healed primarily compared to 79% (53/67) with prior flap coverage. 77% of flaps with one prior flap healed successfully, 79% with two prior flaps and 81% with three or more flaps (Fig. 1). Flap success was not affected by prior flap reconstructions (P = 0.58, Chi squared test). Therefore, although certain flaps are not used as first-line coverage (TFL, gracilis) this does not appear to significantly affect their healing rates.
The average size of the debrided wound prior to reconstruction was 75 cm*. When the size of the defect following debridement was analysed to determine an effect on healing, smaller defects (an average of 59.6 cm) were paradoxically less likely to heal success- fully (P = 0.0089, Chi squared test) (Fig. 2). Healed wounds averaged 82.9 cmZ. It is possible that some other risk factor(s) was related both to the size of the wound and the outcome. It is also possible that this difference may be an artefact of the imbalance in
376 British Journal of Plastic Surgery
P !?! 40
z z h 30
6 $ 20
2 z 10
0 1 2 Number of prior flaps
q Healed q Not healed I
I q Healed q Not healed *P = 0.0089
Pressure sore area (cm2)
Figure l-Prior ischial flap reconstructions vs. healing. Figure 2-&e of debrided ischial pressure sore vs. healing.
sample sizes between those ischial sores that healed Table 3 Risk factors for impaired wound healing versus those that did not heal. Further review of this subset of patients showed that 71% (15/21) of flaps in patients with small wounds had more than one risk factor for impaired wound healing (Table 3).
With an average follow-up of 10.7 months, 17% (19/l 12) of the pressure sores recurred. Although the TFL flap (for ischial coverage) seemed to recur most frequently, the sample size was too small to perform any significant comparative analysis between the vari- ous types of flaps. In addition, the relatively short- term average follow-up precludes any in depth analysis of recurrence rates.
Age >65 years Smoking Diabetes Renal failure Wound infection Systemic infection Local radiation Immunosuppressive agents Haemodynamic instability Prolonged hyperalimentation Low total protein
(List generated from previous wound healing studies.12,)
Ischial pressure sores 311
Pressure sores occur most frequently in the elderly and the chronically ill, two groups that will continue to increase in number into the next century. The surgeon caring for patients with advanced, Grade III (extend- ing into the muscle) to Grade IV (involving the under- lying bone and/or joint) pressure sores must now balance several competing and, at times, contradictory demands. First, patients and their families reasonably expect that these sores will be treated in the most efIicacious and expeditious manner possible, which generally means surgical debridement and flap cover- age. At the same time the surgeon practising in todays cost-effective managed care environment is cognizant that surgical procedures are expensive and, while they are generally more cost-effective than prolonged non-surgical therapy for advanced sores, long-term recurrence rates following surgical procedures can exceed 50%.2
This study is one of the largest modern series of pressure sores involving the ischial area and, to our knowledge, is the largest series of inferior gluteal and gluteus maximus flaps reported. Unlike most previous studies, this series compares the efftcacy of one flap to another in providing reliable coverage of the ischial sore. Patients in this study were not randomized and the flaps selected undoubtedly reflect the biases of the surgeons involved. Nonetheless, this data represents the combined treatment by more than twelve attend- ing surgeons over a 17-year period and should thus be widely applicable.
Encouragingly, the data from this study indicates that most ischial pressure sores can be debrided and flap coverage provided in a single operation. This is true regardless of patient age, the number of wound healing risk factors they possess, the size of their pressure sore and even the number of prior flaps. In fact, contraindications to performing immediate flap reconstruction involve only those patients in whom their ability to physiologically tolerate the operative procedure is in question (i.e. severe cardiopulmonary disease, sepsis, etc.).
In this study, the average length of bedrest follow- ing flap coverage and time of hospitalization was significantly shorter than in previously published reports, which suggests improvement in the overall cost-effectiveness of this approach. For example, in 1994 Rajacic et al. reported the most recent notable series of inferior gluteus maximus island flaps for ischial coverage (31 flaps in 27 patients). Typical of other reports in the literature, their standard length of time for...