Advertisement

AAHKS Best Podium Presentation Research Award: Femoral Perforation During Direct Anterior Approach Total Hip Arthroplasty: Incidence, Cohort Characteristics, and Management

Published:March 31, 2022DOI:https://doi.org/10.1016/j.arth.2022.02.110

      Abstract

      Background

      Cortical perforation during femoral preparation is a recognized complication of total hip arthroplasty (THA) but the incidence, patient characteristics, management, and outcome have not been described for the direct anterior approach (DAA).

      Methods

      A database query of all primary and conversion DAA THAs performed by a single surgeon from 2009 to 2021 was used to identify hips that sustained a recognized intraoperative femoral perforation. Radiographs were used to assess stem subsidence, Dorr femur type, and Canal Flare Index.

      Results

      Among 3,973 THAs, 16 patients (0.4%) sustained perforations during broaching including 8 males and 8 females with a mean age at surgery of 65.6 (range 41-81) years and a mean body mass index of 31.0 (range 19.0-44.4). Two hips were converted to longer primary cementless stems to bypass the perforation. For the remaining 14, the broach was redirected and the same primary stem was implanted. Limited weight-bearing (6 patients) or protected weight-bearing as tolerated with a walker/cane (10 patients) was advised postoperatively. At a mean follow-up of 18.7 (range 4-105) months, all stems were stable with no fractures, subsidence, or revisions. Factors associated with perforation included difficulty with exposure (body mass index >40 or a contracture), measurable osteoporosis (Canal Flare Index <3.0), and abnormal proximal femoral anatomy due to prior trauma, retained hardware, or Perthes disease.

      Conclusion

      In this case series, isolated perforation of an otherwise intact femur during DAA was successfully managed with redirection of the broach, implantation of a primary stem that achieved axial/rotational stability, and protected weight-bearing.

      Level of Evidence

      IV, Case Series.

      Keywords

      The use of the direct anterior approach (DAA) for primary total hip arthroplasty (THA) has progressively increased over the last decade [
      • Patel N.N.
      • Shah J.A.
      • Erens G.A.
      Current trends in clinical practice for the direct anterior approach total hip arthroplasty.
      ]. Reported benefits compared to the posterior approach include decreased dislocation rates [
      • Sheth D.
      • Cafri G.
      • Inacio M.C.
      • Paxton E.W.
      • Namba R.S.
      Anterior and anterolateral approaches for THA are associated with lower dislocation risk without higher revision risk.
      ] and accelerated functional recovery in the early postoperative period [
      • Taunton M.J.
      • Mason J.B.
      • Odum S.M.
      • Springer B.D.
      Direct anterior total hip arthroplasty yields more rapid voluntary cessation of all walking aids: a prospective, randomized clinical trial.
      ,
      • Nakata K.
      • Nishikawa M.
      • Yamamoto K.
      • Hirota S.
      • Yoshikawa H.
      A clinical comparative study of the direct anterior with mini-posterior approach: two consecutive series.
      ]. However, the DAA also introduces a unique set of technical challenges, most notably during exposure and instrumentation of the femur.
      One recognized complication of THA performed using an anterior-based approach is perforation of the cortex during preparation of the femoral canal [
      • Talab Y.A.
      • States J.D.
      • Evarts C.M.
      Femoral shaft perforation: a complication of total hip reconstruction.
      ,
      • Wade F.A.
      • Parvizi J.
      • Sharkey P.F.
      • Purtill J.J.
      • Hozack W.J.
      Femoral perforation complicating contemporary uncemented hip arthroplasty.
      ]. Early studies reported the incidence of this complication during cemented primary THA, with incidences ranging from 0.3% to 4.4% [
      • Talab Y.A.
      • States J.D.
      • Evarts C.M.
      Femoral shaft perforation: a complication of total hip reconstruction.
      ,
      • Pellicci P.M.
      • Inglis A.E.
      • Salvati E.A.
      Perforation of the femoral shaft during total hip replacement.
      ]. However, technical notes on the prevention and management of intraoperative cortical perforation have been sparse. Doyle et al [
      • Doyle J.
      • Proctor P.
      • Bessel T.
      • Moloney M.A.
      The mechanical effects of femoral shaft perforation at total hip replacement.
      ,
      • Doyle J.
      • Procter P.
      • Moloney M.A.
      Femoral shaft perforation at arthroplasty: to treat or not to treat.
      ] reviewed the biomechanical implications of intraoperative perforation and Wade et al [
      • Wade F.A.
      • Parvizi J.
      • Sharkey P.F.
      • Purtill J.J.
      • Hozack W.J.
      Femoral perforation complicating contemporary uncemented hip arthroplasty.
      ] presented a series of 4 perforations with uncemented implants that were identified postoperatively after anterolateral THA. Although other investigations have included small subsets of femoral perforations in broader studies of femoral fracture during DAA surgery [
      • Hartford J.M.
      • Knowles S.B.
      Risk factors for perioperative femoral fractures: cementless femoral implants and the Direct Anterior Approach using a fracture table.
      ,
      • Cohen E.M.
      • Vaughn J.J.
      • Ritterman S.A.
      • Eisenson D.L.
      • Rubin L.E.
      Intraoperative femur fracture risk during primary Direct Anterior Approach cementless total hip arthroplasty with and without a fracture table.
      ], we are not aware of prior reports focused solely on femoral perforations recognized during primary, uncemented THA using the DAA. The aim of this study is to present a series of intraoperative femoral perforations to help determine an incidence for this complication, report on the perioperative management and outcomes, and identify cohort characteristics associated with perforation.

      Methods

      A femoral perforation is defined as an isolated hole in the metadiaphysis of an otherwise intact femur that occurs when the broach is misdirected and penetrates through the cortex, without propagation of a fracture line into the surrounding cortical bone. A review of a prospective institutional database was performed to identify all primary and conversion THAs performed from 2009 to 2021 in which a femoral perforation occurred during the process of femoral preparation. This query included all DAA hip replacement procedures performed by the senior surgeon (W.G.H.). Each procedure was performed through a modified Smith-Peterson interval and utilized a specialized fracture table to assist with femoral exposure. The technique for initial femoral preparation evolved over the time period considered for this review, with routine use of a box osteotome followed by a canal finder to access the canal during the early study period. During the latter years of the study, a broach has been increasingly used to access the canal, only using a canal finder if the initial broach did not progress easily. Fluoroscopy was used in all cases to confirm component position, verify leg lengths, and assess intraoperative complications. The attending surgeon used different primary stems during the study interval including a short, flat-tapered wedge design (Tri-Lock; DePuy Synthes, a Johnson & Johnson Company, Raynham, MA), a collared fully hydroxyapatite-coated design (Corail; DePuy Synthes), and a collared short, triple-tapered design (Actis; DePuy Synthes). No stems that rely on diaphyseal engagement were implanted.
      Demographic data, operative reports, and medical charts were reviewed to determine intraoperative management following perforation, postoperative complications, and postoperative outcome scores. Preoperative radiographs were analyzed, and each femoral canal was assigned a Dorr-type based on the femoral morphology by an orthopedic fellow specializing in adult reconstruction (M.C.K.). The Canal Flare Index (CFI) was measured with digital imaging software (Medstrat Echoes, version 3.2.18.0; Medstrat, Downers Grove, IL) using the method described by Noble et al [
      • Noble P.C.
      • Alexander J.W.
      • Lindahl L.J.
      • Yew D.T.
      • Granberry W.M.
      • Tullos H.S.
      The anatomic basis of femoral component design.
      ]. Two outcome measures were utilized during the study period: Harris Hip Score and the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement. Each score ranges from 0 to 100. Serial postoperative radiographs were analyzed to assess stem subsidence. Descriptive analyses (mean, range, and percentages) for patient demographics were calculated using Microsoft Excel.

      Results

      Among 3,973 consecutive primary and conversion THA procedures performed by a single surgeon using the DAA, 16 cases (0.4%) with an intraoperative femoral perforation were identified (Tables 1 and 2) . The mean follow-up was 18.7 months (range 4-105). The cohort consisted of 8 males and 8 females with a mean age at surgery of 65.6 years (range 41-81) and a mean body mass index (BMI) of 31.0 kg/m2 (range 19.0-44.4) with 3 patients (19%) having a BMI >40. The predominant proximal femoral canal morphology was Dorr type B (11/16, 69%), with 3 type A femurs (19%) and 2 type C (13%). The mean CFI was 3.21 (range 2.72-4.28) with 5 patients (31%) having a CFI below the accepted normal range of 3.0-4.7 defined by Noble et al [
      • Noble P.C.
      • Alexander J.W.
      • Lindahl L.J.
      • Yew D.T.
      • Granberry W.M.
      • Tullos H.S.
      The anatomic basis of femoral component design.
      ]. Perforation events occurred over the entire study interval, and did not appear to be more prevalent during the senior surgeon’s early experience that would have included his learning curve (Table 1). Among the 13 patients with data, the mean postoperative outcome score at most recent follow-up corresponding to a mean of 14.1 months (range 4-31) was 83 (range 57.8-100). Nine patients had a postoperative score >80 (Table 2). One score below 65 was associated with a Workman’s Compensation claim. The outcome score (Harris Hip Score) for case 1 was excluded due to severe medical comorbidities including Wegener’s granulomatosis, dialysis dependence, and failed contralateral THA with subsidence that required wheelchair use at all times. Two additional patients did not have postoperative outcome scores recorded, and despite multiple attempts were unable to be reached to arrange late follow-up.
      Table 1Femoral Perforation Patient Demographics.
      Case #Age at Surgery (y)Year of SurgeryGenderBody Mass Index (kg/m2)Surgical SidePreoperative DiagnosisDorr TypeCanal Flare Index
      1762010M25.8RightPost-traumaticC3.00
      2702011F44.4LeftOAB3.56
      3732011F26.6RightOA, 30° knee flexion contractureB3.18
      4722011M28.2RightOAB3.03
      5592013F30.7LeftOAB2.99
      6742014M31.5RightOAB3.29
      7552014M31.4LeftRA/post-traumaticA3.60
      8662016F21.3RightOAB3.51
      9682016M34.5LeftOAB2.94
      10532017M23.6RightOA with a history of PerthesA4.28
      11492017M41.8RightOAB3.08
      12412017F41.0LeftPost-traumaticC2.72
      13812018F30.2LeftOAB2.99
      14742020F26.4RightOAA3.18
      15682020M39.9RightOAB2.72
      16702020F19.0RightOAB3.21
      M, male; F, female; OA, osteoarthritis; RA, rheumatoid arthritis.
      Table 2Perioperative Management and Postoperative Results.
      Case #Broach SystemInitial Postoperative WB RestrictionsMost Recent Outcome ScoreIdentified Factor Contributing to Perforation
      1Short, flat-tapered wedge
      Stem changed to long, metaphyseal-filling double-wedge stem following recognition of perforation.
      10% WBExcluded due to severe medical comorbiditiesPre-existing hardware (3 cannulated screws in femoral neck), osteoporosis
      2Short, flat-tapered wedge50% WB76.8
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      BMI 44.4
      3Short, flat-tapered wedge50% WB83.8
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      Preoperative 30° knee flexion contracture
      4Short, flat-tapered wedge
      Stem changed to long, metaphyseal-filling double-wedge stem following recognition of perforation.
      50% WB65.8
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      5Collared, fully HA-coated50% WB95.8
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      Canal Flare Index < 3.0
      6Collared, fully HA-coated50% WBNot available
      7Long, metaphyseal-filling double wedgeWBAT with walker57.8
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      Post-traumatic OA with sclerotic changes in femoral canal
      8Short, triple-tapered, collaredWBAT with walker81.8
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      9Short, triple-tapered, collaredWBAT with walker99.8
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      Canal Flare Index < 3.0
      10Short, triple-tapered, collaredWBAT with walker100
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      Perthes disease
      11Short, triple-tapered, collaredWBAT with walker90.0
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      BMI 41.8
      12Short, triple-tapered, collaredWBAT with walker81.8
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      Pre-existing hardware (2 trochanteric screws)

      BMI 41.0

      Canal Flare Index < 3.0
      13Short, triple-tapered, collaredWBAT with walkerNot availableCanal Flare Index < 3.0
      14Short, triple-tapered, collaredWBAT with walker68
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). c = HOOS, JR.
      15Short, triple-tapered, collaredWBAT with walker100
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). c = HOOS, JR.
      Canal Flare Index < 3.0
      16Short, triple-tapered, collaredWBAT with walker81
      During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). c = HOOS, JR.
      Broach system describes broaches used when perforation occurred.
      WB, weight-bearing; WBAT, weight-bearing as tolerated; BMI, body mass index; HA, hydroxyapatite;OA, osteoarthritis.
      a Stem changed to long, metaphyseal-filling double-wedge stem following recognition of perforation.
      b During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). b = HHS.
      c During the study period, our institution transitioned from collecting the Harris Hip Score (HHS) to the Hip Disability and Osteoarthritis Outcome Score, Joint Replacement (HOOS, JR). c = HOOS, JR.
      Of the 16 patients, 13 presented with primary osteoarthritis and 3 presented with secondary post-traumatic osteoarthritis. The first case (Table 2, case 1) had a prior femoral neck fracture associated with severe osteoporosis and renal osteodystrophy which had been treated with 3 cannulated neck screws that were removed at the time of surgery. Case 12 sustained a remote acetabular fracture in conjunction with an automotive collision that occurred approximately 20 years prior to her THA. Two retained screws in the trochanter were removed (Fig. 1) and the sclerotic remnants of the screw tracks in the proximal femur likely contributed to the perforation during broaching.
      Figure thumbnail gr1
      Fig. 1In a hip with pre-existing hardware (case 12), intraoperative fluoroscopy demonstrates posterolateral perforation attributed to broach deflection off sclerotic bone.
      Three patients had notable anatomic abnormalities identified preoperatively. Case 3 had a 30° fixed knee flexion contracture that limited the ability to adequately extend and adduct the leg, a critical maneuver to allow for appropriate broach placement. Case 7 had underlying rheumatoid arthritis, and had sustained a prior femur fracture which was successfully treated with an intramedullary rod that was subsequently removed prior to the patient’s THA. Regions of sclerotic bone had formed in the femoral canal as a result of the prior hardware, which was appreciated on preoperative radiographs. Case 10 had osteoarthritis associated with severe Perthes disease.
      The planned stem for the initial 4 patients in this case series was a short, flat-wedge taper design. In 2 of these cases, the stem was changed to a longer, metaphyseal-filling double wedge design (cases 1 and 4; Fig. 2). A cerclage wire was placed during case 1 to prevent fracture propagation. In the remaining 2 cases (cases 2 and 3), the original planned stem was reimplanted after redirection of the broach. Subsequently, a fully hydroxyapatite-coated stem was used for several years during which 2 perforations occurred. More recently, a short, triple-tapered, collared stem has been utilized as the primary stem (Fig. 3). There was no change in the ultimate stem choice following the recognition of perforation in any case using either of these implants. For the latter cases in this series, autograft bone from the acetabular reaming was packed into the perforation site from inside the canal prior to stem placement, but this technique was not documented in the database so the number of cases that had this performed is not available.
      Figure thumbnail gr2
      Fig. 2Intraoperative fluoroscopy demonstrates femoral perforation associated with a short, flat-tapered wedge stem (case 4). A subtle finding of the distal tip of the broach overlying the cortex indicates that the stem may not be centered in the canal.
      Figure thumbnail gr3
      Fig. 3A hip with primary osteoarthritis and a Canal Flare Index of 2.99 (A) sustained a perforation in conjunction with the use of a short, triple-tapered broach system (Case 13). The posterolateral perforation is depicted by the position of a canal finder (B).
      Over the study interval, there was an evolution in the postoperative management of weight-bearing status. With the initial perforation, the patient was made 10% weight-bearing (WB) for 4 weeks, then 50% for 4 weeks prior to release to weight-bearing as tolerated (WBAT). The subsequent 5 patients were made 50% WB for 4 weeks and then released to WBAT. The final 10 patients were allowed immediate WBAT but were instructed to use a cane or walker for 4 weeks. All patients were informed of the complication and counseled to avoid falls, high-impact exercises, and other excessive stresses on the leg for the first 4-6 weeks after surgery.
      Assessment of healing at the perforation site was difficult to perform, as many of the lesions could not be visualized even on early postoperative radiographs. However, a review of radiographs at most recent follow-up revealed no evidence of stem subsidence in any case, and there were no periprosthetic fractures in the follow-up period. No patient required return to the operating room. The only postoperative complication was 1 deep venous thrombosis (case 3).

      Discussion

      Despite increasing use of the DAA for THA and an associated growth in knowledge about the benefits and pitfalls of this surgical procedure, femoral perforation during broaching remains a largely undescribed complication. This may be due, in part, to its infrequent occurrence with an incidence of 0.4% in this study. However, it remains important to recognize the presence of a perforation at the time it occurs, as all 4 perforations recognized postoperatively in the small case series presented by Wade et al [
      • Wade F.A.
      • Parvizi J.
      • Sharkey P.F.
      • Purtill J.J.
      • Hozack W.J.
      Femoral perforation complicating contemporary uncemented hip arthroplasty.
      ] required return to the operating room for early femoral stem revision.
      Concern for perforation is first raised if the starter broach encounters difficulty entering the femoral canal. At our institution, the canal is identified with a canal finder and the first starter broach is impacted with care. The broach is guided along the path of least resistance, and forceful impaction is avoided if the broach fails to advance. A broach handle that is oriented in the wrong direction raises suspicion, and if a perforation occurs, the hole can typically be felt with a canal finder, tonsil, or Yankauer suction tip. It should be noted that direct visualization of the hole can be quite difficult depending on the location within the metadiaphyseal region, and therefore the presence of a perforation is instead confirmed by recognizing extension of the broach (or any other radio-opaque instrument) outside the cortex on fluoroscopic images. Multiple fluoroscopic views as well as direct visualization should then be performed to ensure that there is no extension of a fracture line beyond the region of perforation. A direct posterior perforation can be subtle on an anteroposterior image, and we therefore recommend obtaining orthogonal fluoroscopic images in all cases by rotating the femur. When a concern for perforation exists, this eliminates the possibility of an unrecognized perforation.
      If recognized at the time of the perforation, intraoperative management requires redirecting the broach back into the femoral canal and upsizing until a stable stem position is achieved. It has previously been recommended that if a perforation occurs, it should be treated with conversion to a modular, revision-style implant with a fluted stem and a proximal sleeve [
      • Talmo C.T.
      • Bono J.V.
      Preventing and managing intraoperative fractures and perforations in hip arthroplasty.
      ]. However, implantation of these stems can be challenging using the DAA, and the sole report using this type of stem for management of an intraoperative perforation in conjunction with the anterior approach required revision for femoral loosening [
      • Cidambi K.R.
      • Barnett S.L.
      • Mallette P.R.
      • Patel J.J.
      • Nassif N.A.
      • Gorab R.S.
      Impact of femoral stem design on failure after anterior approach total hip arthroplasty.
      ]. The results from the current study indicate that conversion to a revision-style stem may not be necessary, particularly if a collared stem that fills the proximal canal is used. A collared prosthesis was utilized in 11 of 16 cases in our cohort which may have contributed to stem stability. Previous studies have reported a reduced rate of periprosthetic fracture and improved biomechanical stability when a collar is used [
      • Lamb J.N.
      • Baetz J.
      • Messer-Hannemann P.
      • Adekanmbi I.
      • van Duren B.H.
      • Redmond A.
      • et al.
      A calcar collar is protective against early periprosthetic femoral fracture around cementless femoral components in primary total hip arthroplasty: a registry study with biomechanical validation.
      ,
      • Demey G.
      • Fary C.
      • Lustig S.
      • Neyret P.
      • si Selmi T.
      Does a collar improve the immediate stability of uncemented femoral hip stems in total hip arthroplasty? A bilateral comparative cadaver study.
      ]. Our results indicate that combining a collar with a stem design that fills the proximal femur appears to provide adequate resistance to axial and torsional forces, allowing proximal femoral perforations to be treated conservatively like Vancouver type A fractures [
      • Gaski G.E.
      • Scully S.P.
      In brief: classifications in brief: Vancouver classification of postoperative periprosthetic femur fractures.
      ].
      At our institution, weight-bearing as tolerated is now allowed with walker assistance, even after perforation occurs. This is always accompanied by a discussion with the patient—both describing the complication and emphasizing the importance of avoiding a stumble or fall. We believe that this discussion with the patient likely contributes to the patient protecting the hip during healing. No patient in this case series demonstrated component loosening or postoperative fracture, and no reoperations have been required.
      Appreciation of patient-specific characteristics that may predispose to femoral perforation is paramount to protect against this complication. Prior studies of perforation with other surgical approaches have identified obesity, osteoporosis, contractures, prior trauma, prior surgery, and aberrant proximal femoral anatomy as risk factors [
      • Talab Y.A.
      • States J.D.
      • Evarts C.M.
      Femoral shaft perforation: a complication of total hip reconstruction.
      ,
      • Pellicci P.M.
      • Inglis A.E.
      • Salvati E.A.
      Perforation of the femoral shaft during total hip replacement.
      ]. Using computed tomography reconstruction and modeling software, Guild et al [
      • Guild 3rd, G.N.
      • Runner R.P.
      • Rickels T.D.
      • Oldja R.
      • Faizan A.
      Anthropometric computed tomography reconstruction identifies risk factors for cortical perforation in revision total hip arthroplasty.
      ] identified specific anatomic criteria predisposing to cortical perforation in the revision setting including decreased patient height, decreased femoral radius of curvature, isthmus width, and the distance of the isthmus from the greater trochanter. More broadly, however, Wade et al [
      • Wade F.A.
      • Parvizi J.
      • Sharkey P.F.
      • Purtill J.J.
      • Hozack W.J.
      Femoral perforation complicating contemporary uncemented hip arthroplasty.
      ] characterized 2 overarching risk factors: “abnormal proximal femoral anatomy” and “difficulty in exposure.” Results from the current study corroborate these observations. Of the 16 patients who experienced a perforation, 11 shared notable characteristics including difficulty with exposure (BMI >40 or a severe knee flexion contracture that limited hip extension/adduction), abnormal proximal femoral anatomy (Perthes disease or retained hardware from a prior surgery), or measurable osteoporosis (CFI < 3.0). With an obese or muscular individual, or if contractures prevent adequate hip extension/adduction, the tendency is for the broach start point to be pushed laterally in space and anteriorly on the femur, which is externally rotated during stem insertion. If unrecognized, this results in an anterolateral-to-posteromedial vector through the femur during broaching, predisposing to posteromedial perforation. In a patient with prior trauma, prior hardware, or bony abnormalities such as Perthes, the presence of regions of sclerosis in an atypical location can direct the broach that should be aligned with the femur (Fig. 4A) away from the femoral canal (Fig. 4B) and out of the cortex (Fig. 4C). In these situations, confirmation of the location of the femoral canal with a canal finder and liberal use of intraoperative fluoroscopy during the broaching process is recommended.
      Figure thumbnail gr4
      Fig. 4Perforation is related to broach alignment relative to the femur. The direction of the broach should be aligned with the direction of the femur (A). When the direction of the broach and the direction of the femur are misaligned (B), perforation can occur (C).
      Although the occurrence of femoral perforation might be expected to decrease over time as the surgeon’s experience with the DAA increased, this was not observed and the cases were distributed over the entire time interval considered for this study (Table 1). This may be explained, in part, by the surgeon’s involvement in a fellowship training program, as several of these cases occurred while the fellow was broaching. It is well-documented that there is a steep learning curve with the DAA, and individual fellows may not have been familiar with the unique subtleties of broaching from an anterior approach. Although the occurrence of a perforation makes for a shocking intraoperative X-ray and raises significant anxiety for the team, the majority of patients in this case series had good to excellent outcomes based on their patient-reported outcome scores.
      This study has several limitations. The study design is retrospective and there is no control or comparison group. In the absence of a group that would have enabled statistical testing, determining whether the cohort characteristics that we have identified constitute potential risk factors for perforation with the DAA should be confirmed by future studies. Although it is possible that we did not capture all femoral perforations among the 3,973 THAs reviewed for this report, we think it is likely that perforations would have been recognized at the time of surgery because intraoperative fluoroscopic images were routinely obtained. This is reinforced by the fact that postoperative radiographs are obtained at 4-week follow-up at our institution and these images have not revealed a femoral perforation that was unrecognized at the time of surgery. Furthermore, due to the longitudinal nature over which the study was performed, many changes occurred in the surgical technique and postsurgical protocols—only some of which are discussed herein. Additionally, all surgeries in this cohort were performed by a single surgeon using a specialized fracture table. As such, our findings may not be generalizable to DAA surgeries performed by other surgeons or on a regular operating table. Finally, due to the low incidence of femoral perforation, this cases series may not represent the complete spectrum of outcomes following femoral perforation.

      Conclusion

      Femoral perforation is a relatively infrequent intraoperative complication associated with the DAA. Based on our institutional experience, this complication can successfully be managed with redirection of the broach, placement of an axially/rotationally stable primary stem, and protected postoperative weight-bearing. Perforations appear to be more common when access to the femoral canal is challenging (obesity, contractures) or when the proximal femur bone morphology is atypical due to the patient’s native anatomy (small CFI, Perthes disease) or secondary to sclerosis from prior surgery.

      Appendix A. Supplementary Data

      References

        • Patel N.N.
        • Shah J.A.
        • Erens G.A.
        Current trends in clinical practice for the direct anterior approach total hip arthroplasty.
        J Arthroplasty. 2019; 34: 1987-1993.e3https://doi.org/10.1016/j.arth.2019.04.025
        • Sheth D.
        • Cafri G.
        • Inacio M.C.
        • Paxton E.W.
        • Namba R.S.
        Anterior and anterolateral approaches for THA are associated with lower dislocation risk without higher revision risk.
        Clin Orthop Relat Res. 2015; 473: 3401-3408https://doi.org/10.1007/s11999-015-4230-0
        • Taunton M.J.
        • Mason J.B.
        • Odum S.M.
        • Springer B.D.
        Direct anterior total hip arthroplasty yields more rapid voluntary cessation of all walking aids: a prospective, randomized clinical trial.
        J Arthroplasty. 2014; 29: 169-172https://doi.org/10.1016/j.arth.2014.03.051
        • Nakata K.
        • Nishikawa M.
        • Yamamoto K.
        • Hirota S.
        • Yoshikawa H.
        A clinical comparative study of the direct anterior with mini-posterior approach: two consecutive series.
        J Arthroplasty. 2009; 24: 698-704https://doi.org/10.1016/j.arth.2008.04.012
        • Talab Y.A.
        • States J.D.
        • Evarts C.M.
        Femoral shaft perforation: a complication of total hip reconstruction.
        Clin Orthop Relat Res. 1979; 141: 158-165
        • Wade F.A.
        • Parvizi J.
        • Sharkey P.F.
        • Purtill J.J.
        • Hozack W.J.
        Femoral perforation complicating contemporary uncemented hip arthroplasty.
        J Arthroplasty. 2006; 21: 452-455https://doi.org/10.1016/j.arth.2005.06.016
        • Pellicci P.M.
        • Inglis A.E.
        • Salvati E.A.
        Perforation of the femoral shaft during total hip replacement.
        J Bone Joint Surg Am. 1980; 62: 234-240
        • Doyle J.
        • Proctor P.
        • Bessel T.
        • Moloney M.A.
        The mechanical effects of femoral shaft perforation at total hip replacement.
        Int Orthop. 1989; 13: 65-68https://doi.org/10.1007/bf00266726
        • Doyle J.
        • Procter P.
        • Moloney M.A.
        Femoral shaft perforation at arthroplasty: to treat or not to treat.
        Arch Orthop Trauma Surg. 1990; 109: 217-220https://doi.org/10.1007/bf00453144
        • Hartford J.M.
        • Knowles S.B.
        Risk factors for perioperative femoral fractures: cementless femoral implants and the Direct Anterior Approach using a fracture table.
        J Arthroplasty. 2016; 31: 2013-2018https://doi.org/10.1016/j.arth.2016.02.045
        • Cohen E.M.
        • Vaughn J.J.
        • Ritterman S.A.
        • Eisenson D.L.
        • Rubin L.E.
        Intraoperative femur fracture risk during primary Direct Anterior Approach cementless total hip arthroplasty with and without a fracture table.
        J Arthroplasty. 2017; 32: 2847-2851https://doi.org/10.1016/j.arth.2017.04.020
        • Noble P.C.
        • Alexander J.W.
        • Lindahl L.J.
        • Yew D.T.
        • Granberry W.M.
        • Tullos H.S.
        The anatomic basis of femoral component design.
        Clin Orthop Relat Res. 1988; 235: 148-165
        • Talmo C.T.
        • Bono J.V.
        Preventing and managing intraoperative fractures and perforations in hip arthroplasty.
        Orthopedics. 2005; 28: s1085-s1088https://doi.org/10.3928/0147-7447-20050902-17
        • Cidambi K.R.
        • Barnett S.L.
        • Mallette P.R.
        • Patel J.J.
        • Nassif N.A.
        • Gorab R.S.
        Impact of femoral stem design on failure after anterior approach total hip arthroplasty.
        J Arthroplasty. 2018; 33: 800-804https://doi.org/10.1016/j.arth.2017.10.023
        • Lamb J.N.
        • Baetz J.
        • Messer-Hannemann P.
        • Adekanmbi I.
        • van Duren B.H.
        • Redmond A.
        • et al.
        A calcar collar is protective against early periprosthetic femoral fracture around cementless femoral components in primary total hip arthroplasty: a registry study with biomechanical validation.
        Bone Joint J. 2019; 101-B: 779-786https://doi.org/10.1302/0301-620X.101B7.BJJ-2018-1422.R1
        • Demey G.
        • Fary C.
        • Lustig S.
        • Neyret P.
        • si Selmi T.
        Does a collar improve the immediate stability of uncemented femoral hip stems in total hip arthroplasty? A bilateral comparative cadaver study.
        J Arthroplasty. 2011; 26: 1549-1555https://doi.org/10.1016/j.arth.2011.03.030
        • Gaski G.E.
        • Scully S.P.
        In brief: classifications in brief: Vancouver classification of postoperative periprosthetic femur fractures.
        Clin Orthop Relat Res. 2011; 469: 1507-1510https://doi.org/10.1007/s11999-010-1532-0
        • Guild 3rd, G.N.
        • Runner R.P.
        • Rickels T.D.
        • Oldja R.
        • Faizan A.
        Anthropometric computed tomography reconstruction identifies risk factors for cortical perforation in revision total hip arthroplasty.
        J Arthroplasty. 2016; 31: 2554-2558https://doi.org/10.1016/j.arth.2016.04.006