Degloving injuries with versus without underlying fracture in a sub-Saharan African tertiary hospital: a prospective observational study

Degloving injury or degloving soft tissue injury has been defined as an avulsion of soft tissue, in which an extensive portion of skin and subcutaneous tissue is detached from the underlying fascia, muscles, or bone surface [1]. These injuries are secondary to shearing forces applied to the tissues, as they happen in road traffic crashes (RTC) [2]. The first reports date back to the early twentieth century, in upper limb injuries caused by occupational accidents with drying machines in laundries, known in the literature as wringer arm [3]. With the advent of the automobile industry, the most frequent trauma mechanism became trampling [2, 4‐6]. There are often high energy involved, involving heavy vehicles with little protection as the case of motorcyclists [2]. This new trend is associated with underlying fractures in 40–85% [2, 4, 7]. They occur as monotrauma or as polytrauma with or without massive blood loss; this has a significance in the estimation of injury severity and the requirement of additional therapeutic options [8, 9]. In degloving injuries, the musculo-cutaneous perforators are ruptured but the skin cover is often viable. Sometimes, traumatic shearing force or crush injury acting on the skin surface can cause a separation of the intact skin and subcutaneous tissue from the underlying fascia; it will create a cavity filled with hematoma and liquefied fat. It commonly occurs over the greater trochanter, but may also occur in the flank and lumbo-dorsal region; this feature is referred to as closed internal degloving, [4, 5, 9, 10]. Its location over the greater trochanter is known as Morel-Lavallé lesion (MLL), as it was first described in the midst nineteenth century [5].

Degloving injuries are managed based on the viability of soft-tissue and the presence of fracture in the degloving-affected areas [1, 7]. This presence of underlying fracture implies a high energy involvement [11]. It is generally accepted that the golden time for avulsing injury treatment is 8 h after injury, because the avulsed skins gradually develop ischemia and necrosis due to circulation disorder [12]. The straightforward treatment in emergency is debridement and repositioning of the avulsed flap back into their original position [1], but it is reported that necrosis of the repositioned flap occurs frequently due to the flap viability prior reposition [2, 13, 14]. Clinical assessment of the viability of soft-tissue envelopes by direct inspection is a weak predictor of the extent of injury [2]; it remains less accurate and challenging in practice [5, 15]. The use of intravenous fluorescein has been proposed as a better assessment method, but may overestimate the line of demarcation between viable and nonviable skin [16]. Generally, after incomplete avulsion, skin color, skin temperature, pressure reaction, and bleeding patterns should be examined carefully to assess tissue viability [17, 18]. Nevertheless, if there is total excision of the avulsed tissue, it will lead to extensive tissue loss, increased morbidity, need for new donor sites, increased number of surgical procedures, and prolonged of hospital stay with increased cost. In addition, if there is underlying fracture, there may be need of reduction and fixation before the soft tissue definitive treatment [2, 7]. Uganda’s growing population and increased motorization, combined with a diverse mix of road users including pedestrians, bicycles, handcarts, motorcycles, cars, busses, trucks, trailers, and animals make the road environment increasingly complex [19, 20]. These predispose the population to an increased risk of complex trauma. Delineating the types of degloving injuries in relation to underlying fracture is the cornerstone of establishing a standardized protocol of their management. The purpose of this study therefore was to compare the outcomes of degloving injuries with and without underlying fracture in view of improving their management in a resource constrained environment.

There were a total of 3279 trauma patients admitted in A & E during a period of 5 months. Among them, we identified 49 patients who presented with degloving injuries. After excluding 3 patients, we recruited 46 patients with a total of 51 degloving injuries into this study as shown in the patients’ flow chart (Fig. 1). Some patients had multiple location of degloving injuries as illustrated in Fig. 2. The prevalence of overall degloving injuries among trauma patients was 1.56% (n = 51); there were 33 (1%) with underlying fracture and 18 (0.56%) without underlying fracture. A total of six patients died and four patients were lost to follow up after discharge.

The majority of the patients were male in 67%; male-female ratio was 2:1. Underlying fractures were more frequent among female patients in 28% with a significant difference (p = 0.0487). The mean age was 28.8 ± 12.8 years without significant difference between the two groups (Table 1).

Table 1 shows also that RTC was the leading cause of injury (84%) resulting in an increased number of degloving injuries with underlying fracture (59%) with significant difference (p = 0.02). Car accident (as pedestrian) or motorcycle accident (as a passenger) were observed in 63% of degloving injuries.

The most common trauma mechanism was knocking (45%), followed by trampling which lead to degloving injury with underlying fracture in 33% (p = 0.0001). The mean ISS was greater in the group of degloving injuries with underlying fracture (13.83 ± 6.25 versus 27.18 ± 21.02; p = 0.0117). The ISS above 25 was associated with underlying fracture. There were 29.41% who presented with shock at admission without significant difference between the 2 groups of degloving injuries (p = 0.2025).

Underlying fractures of both the tibia (19.6%) and fibula (13.7%) were the most frequent, followed by fracture of foot bones (tarsal, metatarsal, and phalanx) and pelvic bones as shown in Fig. 3.

Table 2 shows that the most frequent anatomical location of degloving injuries was the lower extremity (56.14%) followed by the trunk (19.61%). Degloving injuries of the lower limbs tended to be associated with underlying fracture in 45.10% with a significant difference (p = 0.0018), and degloving injuries of the trunk in reverse were not associated with underlying fracture in 13.72% with a significant difference (p = 0.0228). About 96% were open degloving injuries. The mean absolute DBS was 476.10 ± 673.72 cm². The median absolute DBS was 204 cm² without significant difference between the 2 groups of comparison (p = 0.1591). The mean relative DBS in percentage of total body surface area (TBSA) was 3.27%, with median relative DBS of 1.50% without significant difference (p = 0.0777).

Figure 4 shows that the majority of patients with degloving injuries reached the hospital within 120 min from the injury time (71.73%), and the culminant interval of time is between 90 to 120 min from the injury time (21.73%). Table 3 shows that the majority of patients reached the hospital without any treatment of the wound (60.78%); about 33.33% of patients benefited from wound bandage coverage prior to be transferred to the A & E. The majority received analgesics (98.04%), antibiotics (92.16%), and intravenous fluids (86.27%) at the admission in A & E. Table 3 shows that serial dressing was the most common therapeutic practice of overall degloving injuries, followed by primary debridement and closure of the avulsed flap (respectively 76.47 and 37.25%); Fig. 5 illustrates an example of primary debridement and closure in our setting. Patients received routinely antibiotics and analgesics during the definitive management. There were four patients who benefited from diverting colostomies (7.84%) in the group of degloving injuries with underlying fracture of the hip region; an illustration of a case of unstable pelvic fracture with left thigh MLL in a patient who benefited from a diverting colostomy is available in Additional file 1.

The short-term outcomes were good in 62.5% for the overall degloving injuries; there was a significant difference in favor of the degloving injuries without underlying fracture in 32.5% (p = 0.0197). The significance was markedly found in primary healing outcome with p = 0.018 (Table 4).

Degloving injuries with underlying fractures had 3.9 times increased risk of developing poor short-term outcomes (mainly infection) comparing to those without underlying fracture (95% CI 1.02–14.96; p = 0.0197; Table 4). Those infection patterns were especially local skin infection, persistent ulcer, osteomyelitis, and sepsis as illustrated in Figs. 6 and 7.

Overall mortality was 13.04%, without significant difference between the compared two groups. The causes of death were hemorrhagic shock (4.35%), severe head injury (4.35%), and sepsis (4.35%). Further mortality analysis shows that patients who arrived at the hospital after 1 h from injury had 2.96 times increased risk of death (95% CI 1.32–6.64; p = 0.045). In addition, the risk of death was increased in 6.67 times among patients who presented with hemorrhagic shock at admission (95% CI 3.19–13.94, p ˂ 0.0001), and in 5.71 times among patients with the ISS above 24 (95% CI: 2.92–11.20, p = 0.0002). There was also a significant difference (p = 0.404) between the means of relative DBS (in % of TBSA) of patients who died (4.33%) versus those who survived (2.96%).

The mean of hospital stay in days was 21.0 ± 27.12 days. The degloving injuries with underlying fracture had longer hospital stay (26.52 ± 31.31 days) with significant difference (p = 0.0472) as shown in Table 4.

We set out to explore the outcomes of degloving injuries with and without underlying fracture. We reported a prevalence of 1.56% of degloving injuries among trauma patients. Two thirds of those degloving injuries had underlying fracture. This is significant in this setting where there was a high volume of 3479 admitted trauma patients in 5 months without any natural disaster or war.

Degloving injuries occurred more frequently in young males, since they are the most exposed to trauma in everyday life. The overall mean age was 28.8 years; this stands with most of studies where the mean age was between 29 and 31 years [1, 7, 24]. The male-female sex ratio was 2:1; Mello et al. found a similar ratio in Brazil [7]. In our study, females were more frequently affected by degloving injuries with underlying fracture but p value was close to 0.05.

Our study found that RTC was the principal cause of overall degloving injuries in 84.32%. This stands with the findings of other studies done on degloving injuries where the RTC-related injury ranges from 45 to 97% [1, 7, 25]. Pedestrians (involved in car accident) or passengers (involved in motorcycle accident) were both frequently observed in RTC-related causes of degloving injuries. We also found that RTC were frequently associated with degloving injuries with underlying fracture (58.82%) with significant difference (p = 0.0228). Indeed, the presence of heavy trucks on the roads of Kampala city predisposed the population to degloving injuries with underlying fractures (19.61%, p = 0.0092). The most common trauma mechanism of overall degloving injuries was collision or knocking (45.10%), followed by trampling (33.33%). The trampling mechanism resulted significantly to underlying fractures (p = 0.0001). Available literatures described causes rather than trauma mechanisms, yet both have shown significance for the determinant of underlying fracture in degloving injuries. Although we emphasize their impact separately in the occurrence of underlying fracture, those trauma mechanisms may occur concurrently during RTC.

The mean ISS was 22.47 ± 18.38 in overall degloving injuries. The mean ISS was greater in the group of degloving injuries with underlying fracture in comparison to the group without fracture (27.18 ± 21.02 versus 13.83 ± 6.25, p = 0.0117). In contrast, Hakim et al. in Qatar found that the mean ISS was 13.80 ± 10.9 for degloving injuries [24]. These can be explained by the same fact that our patients sustained high energy trauma (knocking and trampling) with associated injuries. The ISS above 25 was associated exclusively associated with underlying fracture in our study.

About 29.41% of degloving injuries resulted in hemorrhagic shock without significant difference regarding to the presence or absence of underlying fracture; degloving injuries are known to be associated with severe concomitant injuries and massive blood loss [8]. The location of degloving injuries such as the scalp, upper limb, and heel may cause significant blood loss with hemorrhagic shock [14, 15, 26]. Yu Chen et al. found that 29.63% of patients with degloving injuries had hemorrhagic shock at admission. Milcheski et al. in Brazil found 9.5% patients with hemodynamic instability at admission; the contrast of their lower percentage may be explained by their inclusion criteria of unstable patients with several parameters like multiple trauma, multiple transfusions, and hypothermia [13]. In our study, we considered only low blood pressure values at the admission as probable result of severe hemorrhage.

The degloving injuries with underlying fracture were 64.71%. The most frequent underlying fractures were the tibia (19.6%) and fibula (13.73%). Mello et al. found 70% of patients with degloving injuries presented with associated fractures, but they included also fractures of the non-degloving areas. Lower limbs (56.14%) followed by the trunk (21.57%) were the most frequent location of degloving injuries as found also in most of studies on degloving injuries of the limbs and trunk but ranging from 44 to 96% [8, 19, 24, 26]. The primary explanation of this frequency is that the lower limbs and trunk represent more than 70% of the TBSA in general. Degloving injuries of the lower limbs tended to be associated with underlying fracture in 50.98% with significant difference (p = 0.0023), and degloving injuries of the trunk in reverse were not associated with underlying fracture in 13.72% (p = 0.0369). This can be explained by the position of the patient during the injury impact.

The mean of DBS was 1.50%, ranging from 0.4 to 18% of TBSA and from 50 to 3200 cm² in absolute size. There was no significant difference of DBS regarding the presence or absence of underlying fracture. Mello et al. in Brazil found an average DBS of 8.2 ± 4.5% with a range of 3–22% [7]. Yu Chen et al. found a range of 150–1500cm² in absolute size [12]. Those differences could be resulting in random error due to variation on wound measurement techniques.

There were almost 96.08% of open degloving injuries; Khan et al. found a similar result of 94% of open degloving injuries; in contrast, Hakim et al. found a lower frequency of 79.78%; this could be explained by the fact that they excluded degloving injuries in which the skin was completely detached.

About 71.73% of patients with degloving injuries reached the hospital within 120 min from the injury time. About 6.52% reached the hospital after 4 h from the time of injury. In fact, most of the injuries secondary to RTC occurred in the highway of Kampala city, and the patient’s evacuation should have passed through a traffic jam. In China, Yu Chen et al. reported in their study that patients reached the hospital within 3.5 h (range 1.5–10 h) after injury [12].

Serial dressing was the most common surgical definitive management (76.47%), followed by primary debridement and closure of the avulsed flap (37.25%) for the overall degloving injuries.

In case of presence of underlying fracture, serial debridement and excision of the avulsed flap were commonly performed with significant difference (p = 0.0373 and 0.0425). This frequent excision of avulsed flap was a potential loss of tissue; it assumed there was poor assessment of flap viability; this excised flap could have still worked temporarily as a biological dressing. Kudsk et al. suggested that local treatments of degloving injuries should consist of evaluating the viability of the flaps, debridement of necrotic tissues, and use of nonviable flap areas as donor of skin grafts in partial full or thickness [2].

Delayed skin graft was done in 19.61% of overall degloving injuries, but most of them were in the group with underlying fracture (17.65%, p = 0.0771). Arnez et al. did 16.5% of skin graft in their series [1]. In most of the studies, delayed skin graft was performed after some days of vacuum-assisted closure (VAC) [1, 12, 13], which was not available for patients in general wards in MNRH during the study period.

There were 5.88% of primary amputations of the limbs. Milcheski et al. reported 28.6% of amputation in their series in Brazil. This high rate seems to be related to their inclusion criteria (only lower limbs) and also can be explained by secondary complications of massive loss of soft tissue [13].

Good short-term outcomes (primary healing, satisfied skin graft take, secondary healing) were observed for the overall degloving injuries in 62.5%; but when comparing the 2 groups, there was a significant difference in favor of the degloving injuries without underlying fracture in 32.50% (p = 0.0197). This significance of good outcomes was markedly found in primary healing which was frequent in absence of underlying fracture (p = 0.018). Indeed, 6 out of 10 of overall skin grafts were declared satisfied, yet there was no use of VAC (not available) prior skin grafting as described in many studies [1, 12, 13]. In contrast, there was 3.9 times increased risk of developing poor short-term outcomes in the group with underlying fractures (p = 0.0197) (Table 4). Lafiti et al. reported that complications of degloving injuries depend on their mechanism, anatomic region, and type and concomitant injuries [8]. In our study, this increased risk could be explained by high energy tissue damage responsible of fractures, and also accessory manipulations by the orthopedic team with meticulous debridement of the soft tissues before closure.

Mortality was 13.04%, without significant difference between the compared two groups. This is basically an in-hospital mortality of degloving injuries; less probably, more severe polytrauma patients with pelvic fracture would have reached the hospital within 2 h of injury. The reported causes of death in the hospital were hemorrhagic shock (4.35%), severe head injury (4.35%), and sepsis (4.35%). Hakim et al. observed an overall mortality of 9.0% (n = 16) among patients with degloving injuries, and around half (n = 7) of them died within the first 24 h of admission due to severe associated injuries [24]. In our study, we also found that patients admitted after 1 h from the injury time and patients with hemorrhagic shock had increased risk ratio of death, respectively, RR = 2.96 (95% CI 1.32–6.64; p = 0.045) and RR = 6.67 (95% CI 3.19–13.94, p ˂ 0.0001). Those are commonly reported causes of death of trauma patients in our setting. Indeed, ISS above 24 was found to be associated with mortality of degloving injuries in our study (RR 5.71; 95% CI 2.92–11.20, p = 0.0002). Although there is new accurate anatomical score (NISS) to predict mortality in trauma [27], ISS can still be used in degloving injuries to predict mortality. Hakim et al. found that ISS was a mortality predictor of degloving injuries in closed types only (odd ratio 1.2; 95% CI 1.06–1.35; p = 0.004); however, this study was a retrospective series [24]. Larger sizes of the degloving wounds have been shown to be associated with mortality in our study; this could be explained by the bleeding patterns and by the increased risk of infections.

The mean of hospital stay was 21.0 ± 27.12 days (range 0.2–107 days) for the overall injuries. Degloving injuries with underlying fracture had longer hospital stay (26.52 ± 31.31 days) with significant difference (p = 0.0472). In fact, fracture by itself may lead to increased hospital length of stay in regard to eventual manipulations of the wound. Most of the studies gave an average of hospital stay similar to our findings for the overall degloving injuries; Hakim et al. found in their study a hospital length of stay of days 10 (range 1–393) [8] while Mello reported a mean hospital stay of 47.3 ± 40 days (range 7–239) [7]; Milcheski et al. reported a mean hospital stay of 46.2 days for patients undergoing primary suture and 32.5 days for patients undergoing primary grafting (p < 0.001) [13].

The overall degloving injuries occur more frequently in young males and caused by RTC, by collision and trampling mechanism. Severity of injury increases the presence of underlying fracture. The lower limb is the frequent anatomical location of degloving injuries, and associated with eventual underlying fracture. Serial debridement and excision of avulsed flaps were the most performed surgical treatments of degloving injuries with underlying fracture in MNRH. The risk of poor short-term outcomes (infections) and longer hospital stay is increased with the presence of underlying fractures. Our study did not find an association of underlying fracture with the in-mortality among patients with degloving injuries. However, factors such as hospital admission more than 1 h from the injury, presence of hemorrhagic shock at admission, ISS above 24, and larger DBS were significantly associated with the mortality of degloving injuries.