Operative Treatment of Displaced Pipkin Type I and II Femoral Head Fractures

Arch Orthop Trauma Surg (2014) 134:637–644 DOI 10.1007/s00402-014-1960-5

Trauma Surgery

Operative treatment of displaced Pipkin type I and II femoral head fractures

Mohamed Fathy Mostafa · Wael El-Adl · Mostafa Abd-Elkalek El-Sayed

Received: 13 June 2013 / Published online: 26 February 2014 © Springer-Verlag Berlin Heidelberg 2014

Abstract one patient (8.1 %) of trochanteric flip-approached group Background and purpose The optimal surgical approach and two (18.1 %) of posterior approach group. All patients for the treatment of femoral head fracture remains con- were followed up for an average of 31 months (range 24– troversial. We retrospectively reviewed patients with Pip- 84). Except for one patient, the final outcomes were equal kin type I and II femoral head fractures managed surgi- in the two groups. cally through posterior Kocher–Langenbeck approach and Conclusions Despite the limited number of patients, we Ganz trochanteric flip approach aiming to define the best can conclude that good final outcome does not necessarily approach with the least morbidity. follow a specific approach. Patients and methods Between May 1995 and Novem- ber 2010, 23 patients (14 men and 9 women) with an Keywords Femoral head · Pipkin · Trochanteric flip · average age of 39.1 years (range 27–62) were treated by Kocher–Langenbeck open reduction and internal fixation of femoral head fractures (5, Pipkin type I; and 18, type II) through Kocher– Langenbeck approach in 11 patients and trochanteric flip Introduction (digastric) osteotomy in 12. The two approaches were compared regarding operative time, difficulty of reduction and Fracture of the femoral head represents a severe injury to fixation, amount of blood loss, occurrence of femoral head the hip joint and historically has been associated with a rel- osteonecrosis or heterotopic ossification and the final func- atively poor functional outcome [1]. About 5–15 % of pos- tional outcome. The scale of Brooker was used to document terior hip dislocations have been reported to be associated heterotopic ossification. The modified Merle d’Aubigne with femoral head fractures [1–3]. The most widely used and Postel as well as Thompson and Epstein scores were classification was that of Pipkin [4] which is based on the used for final evaluation. location of the femoral head fracture in relation to the fovea Results Trochanteric flip approach was associated with and additional lesion on the femoral neck or acetabulum less operative time, less blood loss and improved visu- (Fig. 1). alization facilitating direct screw fixation compared with The first description of femoral head fracture following Kocher–Langenbeck approach. Non-union of the trochan- posterior hip dislocation was by Birkett [5]. Several case teric osteotomy developed in one patient. Heterotopic series have subsequently been published, however, lack of ossification was seen more in trochanteric flip-approached uniformly applied classification, limitations in the study cases. Avascular necrosis of the femoral head occurred in design and insufficient length of follow-up make it difficult to establish any firm conclusions regarding the optimal treatment of femoral head fracture. There is still no M. F. Mostafa (*) · W. El‑Adl · M. A.-E. El‑Sayed consensus on the management of Pipkin type I and type II Department of Orthopedic Surgery, Mansoura University injuries: whether to treat these fractures operatively or non- Hospital, 36 Al‑Gomhoria Street, P.O. Box 35516, Mansoura, Egypt operatively, whether to fix or excise the head fragment, or e-mail: [email protected] which surgical approach to use [6–8]. Although satisfactory

1 3 638 Arch Orthop Trauma Surg (2014) 134:637–644

Fig. 1 The Pipkin classification. Type I femoral head fracture inferior ral neck fracture. Type IV any femoral head fracture with an associ- to the fovea centralis. Type II fracture extended superior to the fovea ated acetabular fracture centralis. Type III any femoral head fracture with an associated femo- results have been reported with closed or open reduction through posterior Kocher–Langenbeck approach or Ganz and traction treatment, most currently studies recommend trochanteric flip approach and had complete radiographic open reduction and internal fixation of displaced femoral and clinical follow-up for a minimum of two years. Patients head fractures with well-recessed cancellous or Herbert treated through anterior Smith–Petersen approach or ante- screws [9–11]. rolateral Watson–Jones approach were excluded from this The effect of the surgical approach on postoperative het- study. By February 2004 we popularized the trochanteric erotopic ossification (HO) and avascular necrosis (AVN) of flip approach for surgical management of all types of dis- the femoral head is unclear. Orthopedic surgeons continue placed femoral head fractures except cases of Pipkin type to discuss which approach is better for femoral head frac- IV with low posterior wall fractures. tures. Anterior approach was contraindicated by Epstein A total of 23 patients was available, 11 were treated et al. [1]. They noted that most of the blood supply is dam- through Kocher–Langenbeck approach (group A) and 12 aged with posterior dislocation, and a surgical procedure through trochanteric flip approach (group B). There were that embarrasses whatever blood supply remains is there- 14 males and 9 females with an average age of 39.1 years fore inadvisable. Heterotopic ossification has been noted (range 27–62 years) at the time of injury. According to the to occur with a higher incidence in patients who undergo Pipkin classification, five fractures were classified as type an anterior approach [12, 13]. Recently a posterior-based I and 18 as type II. All fractures were associated with pos- approach with a trochanteric-flip osteotomy and a surgi- terior hip dislocation and secondary to motor vehicle acci- cal hip dislocation has been advocated for the management dents in 16 patients, motorcycle accidents in 5, and fall of the femoral head fractures [14, 15]. This approach pre- from height in 2. Associated injuries were present in eight serves the deep branch of the medial femoral circumflex patients (three head injuries, five multiple other fractures) artery (MFCA) and at the same time allows unimpaired with an average injury severity score of 19 (range 10–30). complete visibility of the femoral head. All patients presented to our emergency department and The current study was conducted to compare two surgi- after initial evaluation according to the Adult Trauma Life cal approaches (Kocher–Langenbeck approach and Ganz Support (ATLS™) guidelines, immediate closed reduction trochanteric flip osteotomy approach) used for the sur- of the hip fracture dislocation was attempted under gen- gical treatment of Pipkin type I and type II femoral head eral anaesthesia. Sufficient muscle relaxation was essential fractures over a 12-year period aiming to ascertain which to prevent further cartilage damage as well as iatrogenic approach is safe and reliable. femoral neck fracture. Failed closed reduction (one case) indicated open reduction and internal fixation. Follow- Patients and methods ing reduction, plain radiographs in two planes and a spiral computed tomography scan were performed to evaluate the We retrospectively reviewed the records and radiographs of quality of reduction. Operative treatment was indicated for all patients with femoral head fractures who were treated fractures displaced 2 mm or more. surgically by open reduction and internal fixation at our institution between May 1995 and November 2010. Inclu- Surgical technique sion criteria were adult patients with Pipkin type I and type II femoral head fractures of greater than 2 mm dis- Surgery was performed with the patients in the lateral placement, underwent open reduction and internal fixation decubitus position on the contralateral side using a standard

1 3 Arch Orthop Trauma Surg (2014) 134:637–644 639

Table 1 Modified Merle Score Pain Ambulation ROM (%) d’Aubigne-postel score [17] 6 No pain Normal 95–100 The overall numeric score is 5 Slight or intermittent No cane, but slight limp 80–94 given by adding the domain 4 Pain after ambulation, but disappears Long distance with cane or crutch 60–79 scores 3 Moderately severe, permits ambulation Limited, even with support 40–59 Clinical grades: Excellent 18, 2 Severe with ambulation Very limited Good 15–17, Fair 12–14, Poor 1 Severe, prevents ambulation Bedridden 0–39 3–11

OR table. Patients were draped in a manner to allow free repaired femoral head was then gently reduced after careful movement of the injured limb. In the posterior Kocher– retraction of the cut edges of the capsule. The capsular tear Langenbeck approached hips (group A), the posterior was repaired, piriformis tendon was reattached and all mus- border of the gluteus medius and minimus was identified cle tears were repaired. and retracted to expose the posterosuperior aspect of the In the group of patients who were treated through tro- hip capsule. This was facilitated by abduction and internal chanteric flip approach (group B), patient’s position and rotation of the limb. The piriformis tendon was tagged and skin incision were the same as in Kocher–Langenbeck released but the short external rotators (gemelli and obtura- approach. The original approach of Ganz et al. [14] was tor internus) were preserved. In most of patients, the pos- strictly followed without modification. The trochanteric terior capsule was injured and the femoral head came out flip osteotomy was carefully performed staying lateral to through the interval between the piriformis and the short the insertion of the short external rotators. Care was taken external rotators injuring the fleshy muscles. The extent of to avoid injury of the acetabular labrum while doing cap- this injury was assessed carefully and protected. Of special sulotomy. The fractured fragment was handled, reduced interest was the potential damage to the obturator externus and stabilized in the same way as in Kocher–Langenbeck muscle which may indicate more severe damage because approach. After gentle reduction of the hip, the capsule was of its close relationship to the medial femoral circumflex repaired but not tightened to avoid tension on the retinacu- artery. The soft tissues covering the short external rotators lar vessels that may lead to a drop in the perfusion of the and quadrates femoris muscles were kept intact. The pos- femoral head. teroinferior most extent of the capsular tear was identified Postoperative pain was managed with patient-controlled and protected with a heavy suture to prevent further inferior analgesia. Compression stockings were applied while extension at the time of surgical dislocation. To prevent dis- administration of low-molecular-weight heparin was not ruption of the subsynovial retinacular vessels, the capsule a routine to avoid deep venous thrombosis. Indometha- was left attached to the femoral neck. The capsular rent was cin in a dose of 75–150 mg/day was given to 11 patients extended directly toward the acetabular rim, and then the who were considered risky for development of HO. These capsule was released off the rim anteriorly or posteriorly as patients had an associated head injury, multiple fractures needed. The hip was slowly and gently dislocated by flex- or marked local muscle damage. No patient received post- ion, adduction and internal rotation. The fractured fragment operative radiotherapy to prevent HO. All patients were of the femoral head which remained in the acetabulum was instructed to continue non-weightbearing for 6–8 weeks removed by sectioning the ligamentum teres and kept in postoperatively. During this period, range-of-motion exer- sterile saline on a side table. Any soft tissue attachments cises of the hip were instituted guided by pain. However, to the bone fragment (especially the inferior retinaculum) hip flexion more than 90°, adduction and internal rotation were preserved and the fragment was kept with the femo- were restricted for approximately two months. Serial radio- ral head as this could provide valuable blood supply to the graphs were obtained immediately postoperative and at 8 fragment. In contrast, the blood supply from the ligamen- and 12 weeks as well as 6, 12, 24 months after surgery and tum teres seems negligible. The fractured head fragment at final follow-up. Once radiographs showed solid heal- was then reduced anatomically and temporarily fixed with ing of the osteotomy progressive weightbearing and active Kirschner wires, which were sequentially replaced with exercises to strengthen the abductor muscles were started. small fragment, partially threaded cancellous screws or The two groups were compared as to operative time, Herbert headless screws (Zimmer, Warsaw, IN). The heads difficulty of reduction and fixation, amount of blood loss, of the screws were countersunk below the cartilage level. occurrence of AVN or HO and the final functional outcome. The more anterior and inferior fractured fragment was fixed The scale of Brooker et al. [16] was used to classify het- with lag screws from posterior and superior aspect of the erotopic ossification when present. The modified Merle femoral head after accurate measurement of the length. The d’Aubigne and Postel (Table 1) [17] as well as Thompson

1 3 640 Arch Orthop Trauma Surg (2014) 134:637–644 and Epstein (Table 2) [18] scores were used for final func- The mean operative time for group A was tional evaluation. Regardless of the achieved hip function, 180 21.1 min and for group B 120 19.7 min ± ± the need for hip arthroplasty was classified as a poor out- (P 0.01). The estimated blood loss was 572 90.5 ml = ± come in patients below the age of 60. Heterotopic ossifi- for group A and 283 124.9 ml for group B (P 0.01). ± = cation was classified as a complication if it interfered with Open reduction and indirect screw fixation from the supe- joint motion. Statistical analysis was performed using rolateral aspect of the femoral head was required in two SPSS version 17.0 for windows (SPSS, Inc., Chicago, IL). patients (Pipkin type II) of group A. This was attributed to Pearson Chi square test, Independent sample t test and the lack of visualisation of the more anterior and medial one-way ANOVA test were used to compare the two treat- head fragment. On the other hand, no difficulties were con- ment groups. A P value 0.05 was considered statistically fronted with direct screw fixation in any patient of group B. ≤ significant. One superficial wound infection occurred in group A that resolved with repeated dressing and systemic anti- biotics. No cases of nerve injury or deep infection were Results detected in either group. All trochanteric osteotomies in group B healed uneventfully except in one patient devel- The mean time to reduction of hip dislocation from admis- oped non-union and limping. This 52-year-old woman sion was 4.7 h (range 0.5–18 h). One patient required open was obese and started early weightbearing despite medi- reduction 18 h after failed closed reduction of an irre- cal advice. She was managed by revision of the fixation ducible dislocation. This patient had a displaced Pipkin 9 months after initial surgery. type II fracture and did open reduction through posterior Heterotopic ossification was found in two patients of Kocher–Langenbeck approach. He developed avascular group A (Brooker type II) and four patients of group B necrosis of the femoral head and served by total hip arthro- (Brooker type I in one, type II in two and type III in one). plasty 20 months after surgery. The mean time to reduction None of the patients with HO was elected to undergo sur- including this patient was significantly long (12 5.3 h) in gical excision of the ectopic bone despite the limited hip ± the three patients who had avascular necrosis of the femoral flexion noted in the patient with type III. The significant head (P 0.001). Most of patients (82.6 %) did surgery effect of indomethacin intake in preventing HO could not = within the first few days (mean 3.2 days) after closed reduc- be detected. However, eight of the 11 patients who received tion. The early surgical intervension was closely related to indomethacin had no HO at the final follow-up (P 0.59). = excellent and good final outcome (P 0.03). On the other hand, development of HO was closely related =

Table 2 Thompson and Epstein evaluation criteria [18] Scale Clinical Radiographic

Excellent No pain Normal relationship between head and acetabulum (all of the following) Full range of hip motion Normal articular cartilaginous space No limp Normal density of femoral head No spur formation No calcification in capsule Good No pain Normal relationship between head and acetabulum Free motion (75 % of normal hip) Minimal narrowing of cartilage space No more than a slight limp Minimal de-ossification Minimal spur formation Minimal capsular calcification Fair Pain but not disabling Moderate narrowing of cartilage space (one or more of the following) Limited motion of hip, no adduction deformity Motting of head, areas of sclerosis and decreased density Moderate limp Moderate spur formation Moderate to severe capsular calcification Depression of subchondral cortex of the femoral head Poor Disabling pain Almost complete obliteration of cartilaginous space (one or more of the following) Marked limitation of motion or adduction deformity Relative increase in density of femoral head Redislocation Subchondral cyst formation Sequestrae formation Gross deformity of femoral head Severe spur formation Acetabular sclerosis

1 3 Arch Orthop Trauma Surg (2014) 134:637–644 641 to the presence of an associated injuries especially head joint. Epstein et al. [1] noted that reduction within 24 h injury (P 0.001). gave better results than late reduction. McMurtry and = Avascular necrosis of the femoral head occurred in two Quaile [21] showed that the joint should be relocated patients (18.1 %) of group A and one patient (8.1 %) of within 6 h otherwise the risk of avascular necrosis of the group B. One of the two patients of group A had irreducible femoral head with resultant early degenerative joint disease dislocation with the femoral head buttonholed through the will increase. Chen et al. [22] reported good results with capsule and impaled against the posterior rim of the ace- surgical reduction within 12 h. In the current study, patients tabulum. The other patient had severe head injury that pre- who developed avascular necrosis of the femoral head had cludes surgery for 8 days and internal fixation was difficult the longest time to reduction (mean, 12 h). This supports through the posterior Kocher–Langenbeck approach. The the necessity for early reduction of hip dislocation, but do only patient who developed avascular necrosis of the femo- not define the critical time after which avascular necrosis ral head in group B had a considerable soft tissue injury of occurs. the hip and associated multiple fractures. The matter of which operative approach should be used All patients in the two groups were followed up for for the surgical treatment of femoral head fractures remains an average time of 31 months (range, 24–84 months). controversial. Three surgical approaches are advocated in Except for one patient, clinical outcomes using the Merle the literature: anterolateral (Watson–Jones) [9, 23], ante- d’Aubigne–Postel and Thompson–Epstein scoring systems rior (Smith Petersen) [8, 19], and posterior (Kocher–Lan- were classified equally in the two groups. Ten out of the genbeck) [24, 25] with their advantages have been subject 12 patients (83.3 %) in group B showed good to excellent of previous studies. Recently, a posterior-based approach results, one fair and one poor using both scoring systems. using a trochanteric flip (digastric) osteotomy was intro- Out of the 11 patients in group A, nine (81.8 %) were rated duced with more favorable results [15, 26]. To the knowl- good to excellent and two poor by Merle d’Aubigne-Postel edge of authors, no case matched study comparing Kocher– score while eight were good to excellent, one fair and two Langenbeck approach versus trochanteric flip approach was poor by Thompson and Epstein score. The patient who was published. Swiontkowski et al. [12] compared anterior ver- rated fair in group A was placed in this category because of sus posterior approach in the treatment of Pipkin type I and loss of hip motion that was functionally limiting. The three II fractures and found that anterior approach caused less patients with poor results, 2 in group A and 1 in group B, blood loss, shorter operative time and better visualization had avascular necrosis of the femoral head that required hip and fixation. Despite the added osteotomy that required fix- arthroplasty to relief pain. ation at the end of surgery, trochanteric flip approach was associated with relatively shorter operative time and less blood loss compared with posterior approach in our study. Discussion This could be explained by the time required to deliver the head through the posterior capsular rent without endanger- Fracture of the femoral head associated with posterior hip ing the deep branch of the MFCA, lack of visualisation and dislocation is an uncommon but severe injury [4]. The frac- difficult fixation of the fractured fragment through poste- ture itself as well as the subsequent complications such as rior approach. Furthermore, it should be disclosed that the heterotopic ossification, avascular necrosis of the femoral experience of the surgeon had grown up with the progress head and osteoarthritis may lead to a restriction in hip func- of the study and this may be related to the reduced operation and permanent disability even in young patients. The tive time in group B. absence of a validated outcome instrument has contributed Orientation of the fracture with more anterior and to the lake of absolute recommendations and indications medial fracture fragment made direct screw fixation dif- for the most appropriate treatment of these injuries [19]. ficult especially in obese patients because of soft tissue Although indications for surgical management are quite obstruction. This pushed the surgeon to do indirect fixation clear, controversies still remain regarding the treatment of from the intact superolateral portion of the femoral head in Pipkin type I and type II femoral head fractures, preferred two patients. On the other hand, trochanteric flip approach surgical approach and whether to perform fixation or exci- provided a generous exposure of the femoral head with no sion of the fracture fragment. The principles of treatment difficulties were encountered with direct screw fixation. include prompt reduction of the associated hip dislocation, Non-union of the trochanteric osteotomy is a well rec- early anatomic reduction, rigid fixation of large fragments, ognized complication of trochanteric flip approach [14]. restoration of hip congruency and stability and removal of Gluteus medius is a powerful muscle with its more vertical small and comminuted intra-articular fragments [20]. and lateral fibers are active in maintaining the height of the No consensus exists on the safe time interval between contralateral hip while its more anterior and horizontal fib- injury and reduction of traumatic dislocation of the hip ers are active in rotating the pelvis to bring the contralateral

1 3 642 Arch Orthop Trauma Surg (2014) 134:637–644

Fig. 2 A 52-year-old woman with Pipkin type II fracture dislocation left femoral head (a), plain radiograph and CT scan after closed reduction revealed malposition of fractured fragment (b–d), she was managed by open reduction and internal fixation through trochanteric flip approach (e), she started early weightbearing against medical advice and developed non-union of the trochanteric osteotomy with loosening and backing out of the fixation screws (f), revision of the fixation was performed 9 months later with complete healing on follow-up (g–i)

leg forward. It remains attached to the osteotomized tro- rotation and proximal migration of the fragment when held chanteric fragment, therefore, the trochanteric flip tends to with screw. displace anterosuperiorly. To avoid disruption of the oste- Avascular necrosis of the femoral head may be caused otomy fixation and subsequent non-union, Dixon et al. [27] by damage to the blood supply at the time of injury or may recommended restriction of weightbearing on the operated be iatrogenic. Whereas initial damage to vascular structures side for 6–8 weeks. We found non-union of the trochan- is certainly beyond the surgeon’s control, care should be teric osteotomy in one patient (9.1 %). This female patient taken during reduction maneuver and surgery to avoid fur- was obese and started weightbearing early. Furthermore, ther damage to the blood supply. Epstein et al. [1] strongly one of the fixation screws did not purchase the far cortex advocated posterior approaches for the surgical manage- with loosening and backing out (Fig. 2). This suggests that ment of femoral head fractures. They postulated that the stable fixation of trochanteric osteotomy with well pur- vascular supply likely would be severely damaged after a chased screws, not only avoiding early weightbearing, is an posterior dislocation and the anterior approach could fur- important factor to prevent non-union. Bastian et al. [28] ther damage the remaining blood supply through the intact introduced a stepped osteotomy to enhance stability and lateral femoral circumflex artery (LFCA). However, ana- facilitate anatomic fixation of the trochanteric fragment. tomical and clinical studies did not support this theory Although execution of the step osteotomy is more demand- [23, 29, 30]. Stannarad et al. [19] were able to show that ing than a flat cut, it permits easier reduction and prevents in comparison with an anterior Smith–Petersen approach,

1 3 Arch Orthop Trauma Surg (2014) 134:637–644 643 the posterior Kocher–Langenbeck approach was associated existing injuries. Despite the limited number of patients with an even 3.2 times higher risk of avascular necrosis we can conclude that good final outcome is not necessar- of the femoral head. On the other hand, Ozcan et al. [25] ily follow a specific surgical approach used for the treat- used the posterior approach for treatment of different types ment of such fractures. Trochanteric flip approach is more of femoral head fractures without any major intraopera- versatile, allows excellent exposure of the femoral head tive complications, avascular necrosis or heterotopic ossi- and direct screw fixation with less incidence of avascular fication. They noted that the experience of the surgeon is necrosis. However, it carries the risk of non-union of the the key point for the success of the operation rather than trochanteric osteotomy. Obviously the Kocher–Langenbeck the approach itself. Henle et al. [15] reported the devel- approach does not provide the exposure needed for the opment of avascular necrosis in 2 out of 12 patients man- anteriorly located fragments. In this regard the trochanteric aged through trochanteric flip (digastric) osteotomy. One flip approach is superior. However, the posterior Kocher– of them had an additional operation for excision of periar- Langenbeck approach may be indicated for irreducible ticular ossification. In the current study, we could not find fracture dislocation and with meticulous dissection, avas- a significant difference between posterior approach and cular necrosis of the femoral head can be avoided. Finally, trochanteric flip approach regarding the incidence of femo- we should remember that the most important reasons to ral head avascular necrosis. Posterior approach was more choose between these two approaches are the type and likely to be followed by avascular necrosis than trochan- location of fracture, concomitant injuries and preference of teric flip approach. However, it remains unclear whether the surgeon. the approach itself, delayed reduction, difficult exposure and osteosynthesis or soft tissue injury caused the development of femoral head osteonecrosis. References Heterotopic ossification developed in two (18.2 %) out of 11 patients treated through posterior approach and four 1. epstein HC, Wiss DA, Cozen L (1985) Posterior fracture-dislo- (33.3 %) out of 12 patients treated through trochanteric flip cation of the hip with fractures of the femoral head. Clin Orthop approach. The relative increased incidence of ectopic bone Relat Res 201:9–17 formation with trochanteric flip approach could be explained 2. Hougaard K, Thomsen PB (1988) Traumatic posterior fracture- dislocation of the hip with fracture of the femoral head or neck, by the excessive stripping of abductors to allow ante- or both. J Bone Jt Surg (Am) 70:233–239 rior mobilization of the trochanteric flip. This seems to be 3. Sahin V, Karakas ES, Aksu S, Atlihan D, Turk CY, Halici M unlikely because in non-traumatic setting the same approach (2003) Traumatic dislocation and fracture-dislocation of the hip: is not associated with significant heterotopic bone formation a long-term follow-up study. J Trauma 54:520–529 4. Pipkin G (1957) Treatment of grade IV fracture-dislocation of the [14]. Furthermore, other risk factors such as traumatic local hip. J Bone Jt Surg (Am) 39:1027–1042 muscle damage, head injury and associated multiple frac- 5. Birkett J (1869) Description of a dislocation of the head of the tures were closely related to the development of HO. In the femur: with remarks. Med Chir Trains 52:133 present study, indomethacin prophylaxis seemed to be effec- 6. Dreinhofer KE, Schwarzkopf SR, Haas NP, Tscherne H (1996) Femur head dislocation fractures. long-term outcome of conserv- tive in preventing the development of HO. Indomethacin or ative and surgical therapy. Unfallchirurg 99(6):400–409 low dose radiation has been considered in patients with rel- 7. Mowery C, Gershuni DH (1986) Fracture dislocation of the evant soft tissue trauma, brain or spinal cord injury or a his- femoral head treated by open reduction and internal fixation. J tory of HO to prevent heterotopic bone formation but with Trauma 26(11):1041–1044 8. Schonweiss T, Wagner S, Mayr E, Ruter A (1999) Late a possible negative effect on fracture healing [31, 32]. No results after fracture of the femoral head. Unfalluchirurg significant difference in the clinical or radiological outcome 102(10):776–783 could be detected between the two groups at the final follow- 9. Dowd GSF, Johnson R (1979) Successful conservative treatment up. In agreement with Ozcan et al. [25], the results were not of a fracture dislocation of the femoral head-a case report. J Bone Jt Surg (Am) 16-A:1244–1246 directly related to treatment modalities. The severity of the 10. Butler JE (1981) Pipkin type-II fractures of the femoral head. J injury, presence of an associated injury, general health of the Bone Jt Surg (Am) 63:1292–1296 patient, timing of reduction of the hip dislocation, and timing 11. Lange RH, Engber WD, Clancy WG (1986) Expanding applica- of the surgery, cartilage injury and subchondral collapse are tion for the herbert scaphoid screw. Orthopedics 9:1393–1397 12. Swiontkowski MF, Thorpe M, Seiler JG, Hansen ST (1992) all important factors that can affect the outcome. Operative management of displaced femoral head fractures: case matched comparison of anterior vesus posterior approaches for Pipkin I and Pipkin II fractures. J Orthop Trauma 6:437–442 Conclusions 13. Mrchetti ME, Steinberg GG, Goumas JM (1996) Intermedi- ate term experience of Pipkin fracture-dislocations of the hip. J Orthop Trauma 10:455–461 Surgical treatment is indicated for displaced Pipkin type 14. Ganz R, Gill TJ, Gautier E, Ganz K, Krugel N, Berlemann U I and II femoral head fractures and must not worsen the (2001) Surgical dislocation of the adult hip: a technique with

1 3 644 Arch Orthop Trauma Surg (2014) 134:637–644

full acess to the femoral head and acetabulum without the risk of 24. Vermeiren JA, van Hoye M (1991) Three cases of femoral head avascular necrosis. J Bone Jt Surg (Br) 81:1119–1124 fracture in a single care accident. J Trauma 31:579–581 15. Henle P, Kloen P, Siebenrock KA (2007) Femoral head inju- 25. Ozcan M, Copuroglu C, Saridogan K (2011) Fractures of the ries: which treatment strategy can be recommended? Injury femoral head: what are the reasons for poor outcome? Uls 38:478–488 Travma Acil Cerrahi Derg 17(1):51–56 16. Brooker AF, Browerman JW, Robinson RA, Riley LH Jr (1973) 26. Kloen P, Siebenrock KA, Raaymaker ELFB, Marti RK, Ganz R Ectopic ossification following total hip replacement: incidence (2002) Femoral head fractures revisited. Eur J Trauma 4:221–223 and a method of classification. J Bone Jt Surg (Am) 55:1629–1632 27. Dxion SM, Reddy RP, Williams D, Fern D, Norton (2010) Non- 17. Matta JM, Mehne DK, Roffi R (1986) Fractures of the ace- union following bilateral simultaneous ganz trochanteric osteot- tabulum. early results of a prospective study. Clin Orthop omy. Orthop Rev (Pavia) 2(1):e1. doi:10.4081/or.2010.e1 205:241–250 28. Bastian JD, Wolf AT, wyss TF, Notzli HP (2009) Stepped oste- 18. Thompson VP, Epstein HC (1951) Traumatic dislocation of the otomy of the trochanter for stable, anatomic fixation. Clin Orthop hip: a survey of two hunderd and four cases covering a period of Relat Res 467:732–738 twenty-one years. J Bone Jt Surg (Am) 33:746–778 29. gautier E, Ganz K, Krugel N, Gill T, Ganz R (2000) Anatomy 19. Stannard JP, Harris HW, Volgas DA, Alonso JE (2000) Functional of the medial femoral circumflex artery and its surgical implica- outcome of patients with femoral head fractures associated with tions. J Bone Jt Surg (Br) 82:679–683 hip dislocation. Clin Orthop Relat Res 377:44–56 30. Khan A, Yates P, Lovering A, Bannister GC, Spencer RF (2007) 20. Droll KP, Broekhuyse H, O’Brien P (2007) Fracture of the femo- The effect of surgical approach on blood flow to the femoral head ral head. J Am Acad Orthop Surg 15:716–727 during resurfacing. J Bone Jt Surg (Br) 89:21–25 21. mcMurtry IA, Quaile A (2001) Closed reduction of the traumati- 31. Moore KD, Gross K, Anglen JO (1998) Indomethacin versus cally dislocated hip: a new technique. Injury 32(2):162–164 radiation therapy for prophylaxis against heterotopic ossification 22. Chen ZW, Zahi WL, Ding ZQ, Lian KJ, Kang LQ, Guo LX, Liu in acetabular fractures: a randomised, prospective study. J Bone Jt H, Lin B (2011) Operative versus nonoperative management of Surg (Br) 80(2):259–263 Pipkin type II fractures associated with posterior hip dislocation. 32. Burd TA, Hughes MS, Anglen JO (2003) Heterotopic ossification Orthopedics 34(5):350. doi:10.3928/01477447-20110317-09 prophylaxis with indomethacin increases the risk of long-bone 23. Stockenhuber N, Schweighofer F, Seibert FJ (1994) Diagnosis, non-union. J Bone Jt Surg (Br) 85(5):700–705 therapy and prognosis of Pipkin fractures (femur head dislocation fractures). Chirurg 65:976–982

1 3 本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。

学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，

提供一站式文献检索和下载服务”的24 小时在线不限IP 图书馆。图书馆致力于便利、促进学习与科研，提供最强文献下载服务。

图书馆导航：

图书馆首页文献云下载图书馆入口外文数据库大全疑难文献辅助工具