2.19 Late complications
1 Avascular necrosis
1.1 Femoral head necrosis
The reported incidence of femoral head avascular necrosis ( Fig 2.19-1 ) (F-AVN) is 2–10% [1–11]. Letournel and Judet [4] reported a 3.8% incidence of F-AVN. Of 22 cases, 17 were associated with a posterior dislocation of the femoral head. The prevalence of F-AVN after a fracture associated with a posterior dislocation was 17 (7.5%) of 227; after an anterior dislocation, 1 (1.5%) of 63; and after a central dislocation, 4 (1.6%) of 243. On further analysis of these cases, Letournel and Judet [4] found no correlation with the time to reduction of the femoral head or the quality of the reduction of the acetabulum. They stated, “In fact, the fate of the femoral head appears to be decided from the outset…” [4]. However, this may not be true for a pure dislocation, which requires an immediate reduction to avoid avascular necrosis (AVN).
Matta [12] reported that 8 (3%) of 262 patients who were operatively treated for acetabular fractures developed F-AVN. Six of the eight had a poor result. Helfet and Schmeling [2] had four patients who developed significant F-AVN. Three had transverse posterior wall acetabular fractures with an associated posterior femoral head dislocation; one of these three was additionally noted to have a fracture of the femoral head. The fourth case was a patient with a T-type acetabular fracture and significant posterior femoral head subluxation. In all four cases, the initial radiological diagnosis was made 3–6 months postoperatively but before progressive clinical deterioration began. At the time the series reported three of the patients had a total hip arthroplasty (THA) and the authors believed that the fourth patient was likely to require one as well. Moed at al [13] presented seven cases of F-AVN in 100 patients after operative treatment of a posterior wall fracture. In five of the seven cases, they found that the reduction of the associated hip dislocation was more than 12 hours. They recommended urgent reduction of the hip to avoid F-AVN. Bhandari et al [14] in a series of 109 acetabular fractures with an associated posterior hip dislocation did not find an association between time to reduction of the hip and outcome. They found that the quality of the reduction did influence outcome.
One of the major problems with F-AVN is establishing the diagnosis. The differential diagnosis of progressive destruction of the femoral head includes: AVN, posttraumatic arthrosis (malreduction), and intraarticular hardware. Traditional methods of establishing the diagnosis of F-AVN include plain x-ray, technetium scintigraphy, or magnetic resonance imaging. None of these can guarantee the diagnosis postoperatively in patients with acetabular fractures. Posttraumatic arthrosis occurred in 10% of Letournel and Judet’s [4] cases, even when anatomical reduction was obtained. Letournel and Judet [4] made the diagnosis of femoral head F-AVN in 86% of the cases within the first 18 months. If x-rays were normal at 18 months, they believed that F-AVN would not occur.
We recommend that the diagnosis of F-AVN of the femoral head be reserved for cases of progressive destruction of the femoral head following anatomical reduction of the acetabulum with no evidence of intraarticular hardware. The diagnosis should be established within the first 18 months after surgery. Other causes of progressive joint space narrowing must be excluded before establishing the diagnosis of F-AVN.
1.2 Necrosis of the acetabulum
The only report in the literature of acetabular avascular necrosis (A-AVN) is by Letournel and Judet [4], who documented three cases of anterior column A-AVN. After critical reevaluation later, two of these cases were found to be the result of intraarticular hardware. The cause of the third case is still in question, but Letournel [4] has labeled this case cartilage necrosis.
Letournel and Judet [4] found three cases of posterior wall AVN associated with F-AVN. Intraarticular hardware was found at revision surgery in all three cases. The question they ask is whether the posterior wall crumbled to expose the hardware, or the F-AVN caused irregular wear of the posterior wall leading to destruction. They suggested there was no answer to this question. Neither Helfet and Schmeling [2] nor Matta [15] found any cases that suggested avascular necrosis of the acetabulum in their respective series. Despite lack of evidence in the literature, surgeons with extensive experience in THA after acetabular fracture frequently find areas of articular collapse, which in some cases may represent AVN of acetabular fragments.
Devascularization of the iliac wing, especially the anterior column, during extended or combined exposures is certainly a concern. Some of the soft-tissue attachments must be preserved when these types of exposures are used. Soft-tissue stripping of the wing should be done sparingly as needed and indirect reduction techniques should be used if possible.
2 Late infection
Late infection following acetabular fracture surgery also is a rare complication. Letournel and Judet [4] reported five cases, three of which were diagnosed within a few months of surgery and the other two several years after surgery. Mayo [16] reported two late infections diagnoses later than 2 weeks postoperatively in a series of 163 operatively treated acetabular fractures. Helfet and Schmeling [2] reported only one case of an immunocompromised man, who underwent open reduction and internal fixation of a posterior wall acetabular fracture. The initial postoperative course was benign. However, he began complaining of increasing pain in his hip joint 4–6 months postoperatively. X-rays showed progressive narrowing of the joint space and, subsequently, some breakdown of the posterior wall. Cultures were obtained from samples acquired through image intensification-guided aspiration of the hip joint. Staphylococcus epidermidis, Enterobacter cloacae, and Citrobacter species were isolated from the culture. The patient required posterior arthrotomy, removal of hardware, drainage, and a prolonged course of antibiotics. Owing to extensive destruction of the hip joint, intraarticular and extraarticular hip fusion subsequently was performed. The patient has had no further sequelae.
3 Nonunion
Nonunions are exceedingly rare because of the rich vascularity of the pelvis. Letournel and Judet [4] reported four cases (prevalence, 0.7%). Two were both-column fractures, one transverse posterior wall fracture, and one anterior column posterior hemitransverse fracture. In three cases the transverse component did not heal, and a second operation was undertaken to achieve union. Of the three, one is well; one patient was fine for 20 years and then had a THA; one is rated only fair after 1 year; and one, who did not require a second operation to achieve union, had a THA at 7 years. Mears et al [8,17] also reported one case (prevalence, 1%) of pseudarthrosis of the acetabulum following surgery. Mohanty et al [18] retrospectively reported on seven acetabular nonunions after operative fixation. Five of the seven patients had a transverse or transverse posterior wall fracture pattern. Five of six patients healed after revision fixation and bone grafting. The authors concluded that unstable fixation during the index procedure was the cause of the nonunion.
Nonunions are best managed by revision of the fixation and bone grafting. This is a difficult procedure because of the complex anatomy and previous surgery. Extensive preoperative planning is required. Surgeons experienced with acetabular fractures best manage these problems.
4 Heterotopic ossification
Heterotopic ossification (HO) is a well-recognized complication of acetabular fracture surgery ( Fig 2.19-2 ). The reported incidence varies from 3–69% [1–11]. The lack of consistent use of a single classification scheme makes this rate difficult to interpret. Some authors [4, 19] report only HO that is “significant”; others [6, 20, 21] report HO only if it resulted in loss of active range of motion >20% of normal. Some use the system developed by Brooker et al [2, 3, 6, 10, 20–23] ( Table 2.19-1 ). The Brooker [22] classification is based on a single AP x-ray of the involved hip. This is inadequate for this 3-D problem. Inclusion in one of the various rating schemes seems somewhat arbitrary. In addition, no distinction is made as to whether this is based on a single AP view or on additional data obtained from the Judet views or computed tomography (CT).
Heterotopic ossification after an acetabular fracture has been shown to be related to extensile surgical exposures, male gender, associated head injury, significant delays to open reduction and internal fixation, the fracture type, the severity of the injury, and preexisting skeletal conditions [1–7, 9, 10, 21, 24–31]. Heterotopic ossification has also been reported to be a frequent complication of combined anterior and posterior approaches but is most common with the extensile exposures, especially those that require greater trochanter osteotomy.
Mears and Rubash [8] reported a 69% prevalence (6% significant) of HO in their series of 100 cases. Ninety-seven patients either had a triradiate or extended iliofemoral (EIF) exposure. Reinert et al [10] reported a prevalence of 25% of Brooker [22] III and IV lesions with the use of their extensile approach and trochanteric osteotomy. Letournel and Judet [4] found an overall rate of 24%; the rate for Brooker III and IV HO was 12%. The incidence varied based on the surgical approach used. The incidence of Brooker III and IV HO for the Kocher-Langenbeck (KL) exposure was 10.5%; ilioinguinal exposure, 2%; extended iliofemoral, 35%; and the combined KL/ilioinguinal exposures, 27%. Some authors [23] reported a 54% rate (8% Brooker III and IV) in their 108 cases treated with combined anterior and posterior approaches. Daum et al [25] described a series where 37 of 38 patients were treated with KL, triradiate, or combined KL/ilioinguinal surgical exposures. Eighteen of 38 patients developed some form of HO. Webb et al [31] reported the results of acetabular fracture reconstruction in head-injured patients. Twenty-two of 23 patients developed HO. There were seven cases of Brooker III and six cases of Brooker IV HO. Matta [12] reported an overall 18% incidence of HO; 9% was significant with a 20% loss of motion. The incidence for the KL approach was 8%; ilioinguinal approach, 2%; and EIF approach, 20%.
Despite aggressive stripping of the iliopsoas muscle from the internal iliac fossa, HO is never significant and is almost never seen inside the pelvis with the ilioinguinal approach. It has been reported in the external iliac fossa with the ilioinguinal approach, probably related to a small extrapelvic extension of the approach for the purpose of applying forceps to the iliac wing for fracture reduction [2, 4]. In conclusion, from these reports it appears that the amount of HO probably is related directly to trauma to the hip abductor musculature, whether traumatic or iatrogenic related to intraoperative exposure.
4.1 Heterotopic ossification prophylaxis
Based on the literature for the prevention of HO following THA, several authors developed series using indomethacin, radiation therapy, or a combination of the two to prevent HO following acetabular fracture surgery. A study [29] reported on the experience using 25 mg indomethacin orally three times a day for 6–12 weeks beginning on postoperative day 1. The authors found the overall incidence of HO to be similar in both the control and the treated groups; however, the rate of Brooker III and IV HO was significantly reduced in the indomethacin-treated group. They also noted no significant complications related to the use of indomethacin. A follow-up study [32] noted that the maximal extent of HO was evident by 6 weeks postoperatively. These authors also reported that in patients receiving indomethacin the grade of HO did not improve following discontinuation of the drug. One author [24] had a patient for whom 25 mg indomethacin was administered orally three times a day for 6 weeks postoperatively. The overall rate of HO was 65% in the control group and 50% in the treated group. The incidence of Brooker III/IV in the control group was 38%, whereas in the treated group, 0%. Only one patient discontinued the medication because of gastritis. Letournel and Judet [4] found the overall incidence of HO to be 25% (Brooker III/IV, 11%) when no prophylaxis was used. The overall incidence fell to 10% (Brooker III/IV, 2%) when indomethacin was used for prophylaxis of HO. Matta and Siebenrock [27] used plain x-rays and volumetric CT analysis to determine the extent of HO following acetabular fracture fixation. Patients were randomized into two groups: indomethacin and no prophylaxis. The surgical exposures in the indomethacin group were: KL, 22; EIF, 14; and ilioinquinal, 25. In the control group the exposures were 15, 6, and 25, respectively. The overall incidence of HO on plain x-rays in the indomethacin group was 47% (Brooker II, III, or IV, 9% for KL and EIF) and 57% in the control group (Brooker II, III, IV, 5% for KL and EIF). These comparisons were not statistically significant. The volumetric CT scan of HO for KL and EIF approaches only was 1.7 cm3 for the indomethacin group and 3.6 cm3 for the control group. These differences were also not significant.
Bosse et al [33], based on their experience using 1,000 rad of radiation therapy (200 rad each day for 5 days beginning postoperative day 3), they found the overall incidence of HO decreased from 90% in the control group to 50% in the treated group. The incidence of Brooker III and IV HO also decreased from 50% in the control group to 11% in the treated group. They reported no complications that could be attributed specifically to the radiation therapy protocol. Another study [25] found Brooker III/IV HO in 6 of 24 patients who did not receive radiation therapy and no incidence from 14 patients who had radiation therapy for HO.
One study [28] reported the use of both radiation therapy and indomethacin for the prevention of HO. In this series, the overall incidence of HO was 18%, and there was no incidence of Brooker III/IV HO. The combination of therapies nearly eliminated all HO. Still, the risks involved for both therapies must be weighed carefully against this benefit.
Radiation therapy for HO prevention in acetabular fracture surgery was studied [34], and two protocols were compared: 10 Gy administered in five 2-Gy doses (33 patients) and a single 8-Gy dose (11 patients). An additional three patients received 6–7 Gy in a single dose. Forty-four of the 47 patients began or received their treatment less than 48 hours after surgery. The overall incidence of HO was 24 (51%) of 47. There were six cases of Brooker III HO and no cases of Brooker IV. There was no statistical difference between the two protocols. A single fraction of 7–8 Gy within 24–48 hours of surgery significantly reduced the incidence of clinically significant HO.
A randomized prospective protocol [35] compared radiation therapy and indomethacin for the prevention of HO. There were 33 patients in the radiation group and 39 in the indomethacin group. All HO was present by 6 weeks. The incidence of Brooker III HO was 9% in the radiation group and 13% in the indomethacin group. There was no incidence of Brooker IV HO in the radiation group and 5% in the indomethacin group. There was no significant difference between groups. The authors noted that the two patients in the indomethacin group with Brooker HO failed to receive proper doses of the drug and that radiation therapy was 200 times more expensive. A systemic review [36] showed that radiation was more effective than indomethacin for preventing HO after acetabular fracture surgery.
Potential complications of irradiating the pelvis and its contents include radiation-induced malignant degeneration, sterility, and genetic alterations to sperm or ova. Although these possibilities are exceedingly remote, patients must be well informed of the risks. The complications associated with the use of indomethacin and other nonsteroidal antiinflammatory drugs include the well-known adverse side-effects of the medications, such as gastrointestinal disturbances, coagulopathies, allergic reactions, or the theoretical potential for delayed union of the fracture. Decreased new bone formation, impaired fracture healing, decreased torsional strength of healing bone, and inhibition of haversian bone remodeling in a rabbit model have been reported [29, 32, 37]. It is reported that indomethacin was as effective as placebo in preventing Brooker grades III/IV HO after acetabular fracture surgery with KL approach [38]. However, it was also found that only 57% of patients in the indomethacin group who were tested had detectable levels of serum indomethacin. Noncompliance with therapy could have affected their results.
There is currently no consensus regarding the need for HO prophylaxis after acetabular fracture surgery. Furthermore, if prophylaxis is chosen, there is no agreement regarding the choice of prophylaxis. We currently use 25 mg indomethacin three times per day for 6 weeks or a single 75 mg sustained release capsule per day for 6 weeks. The indomethacin is begun postoperative day 1 for all posterior exposures, all extended exposures, and any anterior exposures, in which extrapelvic placement of a reduction clamp was done. If HO occurs in the first 6 weeks, then indomethacin is continued for a total of 12 weeks after surgery. If indomethacin is not tolerated, is contraindicated, or other risk factors are present, radiation therapy is used instead. Options for radiation prophylaxis are 7–8 Gy in a single dose or 10 Gy in five doses. The radiation therapy began within 24–72 hours after surgery.
Related posts:
2.18 Early complications
2.20 Surgical management of delayed acetabular fractures
2.21 Late acetabular reconstruction
2.23 Results of treatment for fractures of the acetabulum
2.22 Malunion and nonunion
1.8.4 The management of the injured pelvic ring: internal fixation of the anterior pelvic injuries—open book (type B1)
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
