ORIGINAL ARTICLE
Satisfactory Patient-based Outcomes After Surgical Treatment for Idiopathic Clubfoot: Includes Surgeon’s Individualized Technique Susan T. Mahan, MD, MPH,*w Samantha A. Spencer, MD,*w and James R. Kasser, MD*w
Introduction: Treatment of idiopathic clubfoot has shifted towards Ponseti technique, but previously surgical management was standard. Outcomes of surgery have varied, with many authors reporting discouraging results. Our purpose was to evaluate a single surgeon’s series of children with idiopathic clubfoot treated with a la carte posteromedial and lateral releases using the Pediatric Outcomes Data Collection Instrument (PODCI) with a minimum of 2-year follow-up. Methods: A total of 148 patients with idiopathic clubfoot treated surgically by a single surgeon over 15 years were identified, and mailed PODCI questionnaires. Fifty percent of the patients were located and responded, resulting in 74 complete questionnaires. Median age at surgery was 10 months (range, 5.3 to 84.7 mo), male sex 53/74 (71.6%), bilateral surgery 31/74 (41.9%), and average follow-up of 9.7 years. PODCI responses were compared with previously published normal healthy controls using t test for each separate category. Included in the methods is the individual surgeon’s operative technique. Results: In PODCIs where a parent reports for their child or adolescent, there was no difference between our data and the healthy controls in any of the 5 categories. In PODCI where an adolescent self-reports, there was no difference in 4 of 5 categories; significant difference was only found between our data (mean = 95.2; SD = 7.427) and normal controls (mean = 86.3; SD = 12.5) in Happiness Scale (P = 0.0031). Discussion: In this group of idiopathic clubfoot patients, treated with judicious posteromedial release by a single surgeon, primarily when surgery was treatment of choice for clubfoot, patient-based outcomes are not different from their normal healthy peers through childhood and adolescence. While Ponseti treatment has since become the treatment of choice for clubfoot, surgical treatment, in some hands, has led to satisfactory results. Level of Evidence: Level III. Key Words: idiopathic clubfoot, talipes equinovarus, clubfoot treatment
From the *Department of Orthopaedics, Boston Children’s Hospital; and wDepartment of Orthopaedics, Harvard Medical School, Boston, MA. The authors declare no conflicts of interest. Reprints: Susan T. Mahan, MD, MPH, Department of Orthopaedics, Harvard Medical School, 300 Longwood Ave, Boston, MA 02115. E-mail: [email protected]. Copyright r 2014 by Lippincott Williams & Wilkins
J Pediatr Orthop
Volume 34, Number 6, September 2014
(J Pediatr Orthop 2014;34:631–638)
I
n the past decade, treatment of idiopathic clubfoot has shifted from a standard of surgical management to the Ponseti technique.1 Long-term results of the Ponseti technique have been remarkable, which are principally credited to Ponseti himself.2 Long-term surgical outcomes, however, have varied with most authors reporting discouraging outcomes3–6 and some reporting favorable results.7 The majority of comparative studies, albeit retrospective and with shorter follow-up, have favored the Ponseti technique.5,6,8,9 Although Ponseti treatment remains the standard, and has been found to have shortterm success in many different hands,10,11 surgical treatment outcomes varied depending on the treating surgeon and their technique. Although most outcome reports in surgically treated clubfoot discuss results after the Turco technique,3,12,13 more recent use of a judicious, or a la carte posteromedial and lateral release12–14 may result in improved surgical outcomes. Furthermore, “a la carte” surgery is often discussed, but it is not always clear what is on the “menu” and the order of the various procedures. Assessing surgically treated clubfoot can be also difficult, as the proper comparison should be to normal feet, and the assessor should not be the physician but the patient and/or their guardian. Patient-based outcomes comparing children treated for clubfoot to normal healthy controls may be the best assessment of how these feet are functioning. Our purpose was to evaluate a single surgeon’s series of children with idiopathic clubfoot treated with a specific a la carte posteromedial and lateral release using the Pediatric Outcomes Data Collection Instrument (PODCI) with minimum 2-year follow-up. Our primary hypothesis was that children treated surgically for idiopathic clubfoot would be no different in any of the categories of the PODCI when compared with previously published normal healthy controls.15 We also delineate the surgeon’s individual technique.
METHODS After Institutional Review Board approval, we retrospectively identified 148 patients treated surgically by a www.pedorthopaedics.com |
631
Mahan et al
single surgeon (J.R.K.) for idiopathic clubfoot over 15 years, from 1991 through 2006. This was during the time when surgery was the principal treatment option for clubfoot. All 148 patients were diagnosed with idiopathic clubfoot, defined by talipes equinovarus (ICD-9: 754.51) or congenital clubfoot (ICD-9: 754.70) not associated with arthrogryposis, spina bifida, or other conditions known to influence the treatment of clubfoot. All patients’ initial treatment consisted of serial long leg casting but not the Ponseti sequence of casts. Initial clubfoot severity score was not recorded. Tenotomy and foot abduction bracing was not utilized in this group. This treatment was sufficient in approximately 20% of patients and they did not require further surgery. The remaining majority of patients, however, went on to a surgical release, either medial and posterior, or posterior only. See the surgical technique noted below. Patients’ charts were retrospectively reviewed, including operative reports and clinic notes. Patient demographics were noted, including: date of surgery, age at surgery, age at time of analysis, bilaterally of the clubfoot, sex, any complication of the surgery, type of surgery, need for repeat surgery, and recurrence of the clubfoot. Each patient and/or parent was mailed a cover letter, consent form, and an age-appropriate version of the PODCI. An option to complete the instrument online was also provided. If the patient was scheduled for clinic follow-up, the PODCI was made available to them for completion at that time. One hundred forty-eight patients treated surgically for idiopathic clubfoot were mailed a questionnaire. Thirty-one (21%) were returned to sender with no forwarding address. One patient returned the questionnaire indicating that they did not wish to participate. Two patients returned an incomplete questionnaire. Forty patients (27%) did not respond to the questionnaire despite repeat mailings and telephone calls. Seventy-four replied with complete questionnaires, such that 50% of patients were located and included. Patients completed the questionnaire by mail, online, or in clinic (Fig. 1). In the 74 patients who responded to the survey, the median age at surgery was 10 months (range, 5.3 to 84.7 mo), male sex 53/74 (71.6%), bilateral surgery 43/74 (58.1%), and average follow-up at time of survey completion was 9.7 years (minimum 2.0 y, maximum 17.2 y, SD = 4.1 y). In addition, 17.6% (13/74) of the children needed further surgery performed after the initial a la carte posterior only or medial and posterior release. A comparison of patients who did not respond to the survey to those who did respond to the survey was completed to determine if there was any systemic difference between the 2 groups. Initial severity score was not recorded and therefore could not be compared (Table 1). The patients who did not respond to the survey were significantly further out from the surgery temporally (P < 0.005) compared with the responders. There were no other differences between responders and nonresponders in age at surgery, unilateral versus bilateral, sex, or type of surgery (medial and posterior or posterior only) performed.
632 | www.pedorthopaedics.com
J Pediatr Orthop
Volume 34, Number 6, September 2014
Five of the PODCI16 scales were utilized: (1) Transfers and Basic Mobility Scale; (2) Sports/Physical Function Scale; (3) Pain/Comfort Scale; (4) Happiness Scale; and (5) Global Functioning Scale. Patients were divided into the 2 groups as necessary for the PODCI assessment: pediatric (age 2 to 10 y old) and adolescent (age 11 to 18 y old). For the adolescent group, we sent 2 questionnaires: the adolescent self-reported, and the adolescent parent-reported questionnaire. For the pediatric group, the questionnaire was designed to be completed by the parent or guardian. Responses were calculated by the scoring algorithms as created for version 2.0 of the AAOS/COMSS/COSS Outcomes Data Collection Instruments. The standardized score for the PODCI was utilized; all scores are scaled as directed by the PODCI development group from 0 to 100 with the higher score indicating better function or happiness. PODCI responses were compared with previously published normal healthy controls15 using t test for each separate category. We had 39 responses for parent reporting on a pediatric questionnaire (our comparative group had n = 29).15 For adolescent parent-reported PODCI we had 34 responses except for the global functioning score where, because of an incomplete questionnaire, we had only 33 responses (our comparative group had n = 28). For the adolescent self-reported, we had 24 responses (the comparative group had n = 27). We also assessed the responders whose scores fell >2 SD outside the means reported by Haynes et al15 to try and help determine the functional limitations and clinical outcome of these patients. This was a descriptive analysis using their clinic notes and medical records. Additional comparisons were done between patients who had a posterior only versus medial and posterior release. Finally, an assessment of the athletic involvement of these patients was done using question 52 of the PODCI: “Can you/your child participate in competitive level sports with other children the same age (eg, hockey, basketball, soccer, football, baseball, swimming, running [track or cross country], gymnastics or dance)?”
Data Analysis Age at surgery was noted in months (continuous variable). Dichotomous variables, such as sex, bilateral versus unilateral surgery, and need for repeat surgery was noted. Length of follow-up between the surgery and the PODCI assessment was recorded in months (continuous variable). Type of surgery was determined by review of the operative notes, and was a dichotomous variable (posterior and medial release vs. posterior only release). The questionnaires were divided into 3 groups: (1) parent reporting on a child aged 2 to 10 years; (2) parent reporting on an adolescent aged 11 to 18 years; and (3) adolescent aged 11 to 18 years self-reporting. Within each group and scale, a mean score, SD, and sample size was calculated using SPSS (SPSS for Windows, Rel.15.0.0.2006, SPSS Inc., Chicago). These were then compared with student t test assuming different SDs and 2-tailed P-value to previously published normal healthy r
2014 Lippincott Williams & Wilkins
J Pediatr Orthop
Volume 34, Number 6, September 2014
Surgical Treatment of Clubfoot
Contacted patients by mail or clinic
148
Participant
Non-participant
74
74
Respond to mailing
Online
Chose not to participate or didn't complete survey
Return to sender/ no address found
48
5
3
31
Filled out in clinic
No reponse to mailing
21
40
FIGURE 1. Analysis of patients included and not included in the study.
controls15 of similar age, questionnaire group, and scale. This was done using GraphPad version 3.1 (InStat, San Diego, CA). The 2-tailed, unpaired t test, Welch corrected was utilized. It was assumed Gaussian distributions but with populations that may have different SDs. Statistical significance was set a priori at P < 0.05.
Surgical Technique by James R. Kasser, MD The surgical procedure to correct persistent equinovarus deformity of the foot which I used from 1987 until 1998 was rather consistent. Over the first 5 years of practice, to approximately 1987, the procedure evolved becoming more limited. During the evolution, I concentrated the release increasingly on the posterolateral side of the hindfoot, tightened the repair of the heel cord, increased the release of the calcaneocuboid joint from the
medial side, and maintained the spring ligament of the talonavicular joint if possible. Patients had either a posterior only release or a medial and posterior release, depending on the needs of the individual foot. All of the patients in which there was a medial release of the talonavicular joint were pinned antegrade inserting the pin from the posterior aspect of the talus through the head of the talus, centrally through the navicular and exiting between the first and second ray. I did not take x-rays in the OR but confirmed the position of the pin by direct inspection and palpation to be sure that the navicular was neither subluxed dorsally nor laterally. In many cases this meant that there was still some navicular prominence medially after the talonavicular joint was corrected but dorsal lateral subluxation was avoided. Plantar fascia releases were carried out only if there was persistent
TABLE 1. Responders Compared With Nonresponders Analysis Responders N = 74
Nonresponders N = 74
Median age at surgery (mo)
9.63
9.54
Surgery* [n (%)] Unilateral Bilateral Average time from surgery to survey (y)
43 (58) 31 (42) 9.7
38 (58) 27 (42) 11.9
Male sex [n (%)]
54 (73)
53 (71.6)
Type of surgery performed [n/N (%)] Medial and posterior Posterior only
53/74 (72) 21/74 (28)
42/74 (57) 32/74 (43)
Test and P Mann-Whitney U test P > 0.05 Fisher exact P > 0.05 t test P = 0.003 Fisher exact P > 0.05 Fisher exact P > 0.05
*There were 9 nonresponders who had unknown surgical laterality due to unavailability of records.
r
2014 Lippincott Williams & Wilkins
www.pedorthopaedics.com |
633
Mahan et al
flexion deformity through the midfoot. There were no tendon transfers done at the time of initial surgery. The details of the surgical procedure are as follows: The patient was placed in a supine position with a roll under the opposite pelvis. A Cincinnati incision was made, carrying the dissection from the talonavicular joint on the medial side of the foot extending just above the heel crease posteriorly reaching a point beyond the fibula on the lateral side of the foot. The dissection was then carried down through subcutaneous tissue, elevating flaps dorsal and plantar. Sural nerve and lesser saphenous vein were preserved in this dissection. I first entered the peroneal sheath laterally and isolated the peroneal tendons above the level of the distal fibula. I then isolated the heel cord posteriorly and raised a dorsal flap to at least 4 cm. Then the dissection was carried medially, the fascia over the neurovascular bundle was preserved and dissected over the medial malleolus onto the dorsum of the foot. I ensured that the dorsal flap was elevated beyond the midportion of the dorsum of the foot, to allow safe dorsal release of the talonavicular joint. The neurovascular bundle and tendons were preserved in the dorsal layer. The anterior tibial tendon was identified and the dissection carried distally to the most proximal point of its insertion at the first cuneiform. The neurovascular bundle was isolated at the medial malleolus with the dissection carried proximally and distally, defining the medial plantar nerve, the neurovascular bundle, and the calcaneal branch of the nerve. The tendon sheaths of the flexor digitorum longus and posterior tibia were entered with a Z-lengthening of the posterior tibial tendon. In moderate feet, pulleys of tendon sheath were left at the medial malleolus, whereas in severe feet, the tendon sheath was entirely opened across the medial malleolus. In mild feet, the flexor digitorum longus tendon was not lengthened. The posterior tibial was Z-lengthened in all feet. I then opened the talonavicular joint, carrying the dissection first medially, then dorsally, and finally plantarward to the extent necessary to rotate and slide the navicular around the head of the talus. If there was medial subluxation of the cuboid on the calcaneous in severe feet, the medial aspect of the calcaneocuboid joint was entered through the talonavicular joint. The lateral approach to the calcaneocuboid joint and resection of a portion of the cuboid or calcaneus was rarely, if ever, done and only in severe feet. Next, the heel cord was lengthened in a Z-fashion over a distance of 3 to 4 cm. The patient then had the flexor hallucis longus tendon isolated, the tendon sheath opened but the tendon was not lengthened at this point. The extent of opening the tendon sheath was to the posterior aspect of the subtalar joint. One tine of the scissors was then placed through the floor of the tendon sheath of the flexor hallucis longus to open the subtalar joint. This opened the posterior subtalar joint and posterior tibial talar (ankle) joint. I divided the capsule of the subtalar joint to the calcaneofibular ligament. The ankle joint was opened from the flexor hallicus tendon sheath around laterally to the fibula. At this point, generally feet
634 | www.pedorthopaedics.com
J Pediatr Orthop
Volume 34, Number 6, September 2014
will dorsiflex above neutral nicely and that was the extent of the dissection. In occasional cases, more extensive medial capsule was opened. In no cases was the deltoid ligament taken down; in no cases was the interosseous ligament divided. The talonavicular joint was pinned as stated. In severe feet, the tendons of the flexor hallicus and flexor digitorum longus tendon were sewn together proximally and distally and the combined tendon lengthened in a Z-fashion. In mild feet, either the flexor tendons of the toes were not lengthened or lengthened in a fractional way. The Z-lengthening was occasionally done in flexor tendons to the toes but only in older children. The wound was closed as much as possible; the posterior skin was left open if necessary and many feet had a gap as wide as 1.5 cm posteriorly. The neurovascular bundle was covered with soft tissue; a bivavled cast was placed initially in equinus with the foot dorsiflexed under anesthesia at 1 week postoperatively. Again, if the wound remained open, it was simply covered with a nonadherent dressing. The long leg cast remained in place for 2 months; the pin was removed at 1 month. The patient was then started on exercises and occasional postoperative bracing or casting was needed, but in general, simply physical therapy was carried out with no long-term bracing. Many patients did not require the medial release of the talonavicular joint and had a posterior release only. This included Z-lengthening of the Achilles, capsular release of the subtalar and ankle joints, and release of the calcaneofibular ligament. In these patients with generally less involved clubfeet, no pin was placed.
RESULTS There were no statistical differences found among the 5 scales in the pediatric PODCI (aged 2 to 10 y) and the adolescent parent-reported PODCI (aged 11 to 18 y) between our clubfoot patients and normal healthy historical controls. Analysis of the, adolescent self-reported PODCI (aged 11 to 18 y) showed one statistically significant difference in the Happiness Scale, in which our surgically treated clubfoot patients were happier than the normal healthy historical controls (our mean 95.2 and SD = 7.427 compared with historical controls mean 86.3 and SD = 12.5; P = 0.003) (Table 2). In the analysis of the responders who were outside 2 SD of the means reported by Haynes et al,15 there were 22 patients (31.9%) who fell
Satisfactory Patient-based Outcomes After Surgical Treatment for Idiopathic Clubfoot: Includes Surgeon’s Individualized Technique Susan T. Mahan, MD, MPH,*w Samantha A. Spencer, MD,*w and James R. Kasser, MD*w
Introduction: Treatment of idiopathic clubfoot has shifted towards Ponseti technique, but previously surgical management was standard. Outcomes of surgery have varied, with many authors reporting discouraging results. Our purpose was to evaluate a single surgeon’s series of children with idiopathic clubfoot treated with a la carte posteromedial and lateral releases using the Pediatric Outcomes Data Collection Instrument (PODCI) with a minimum of 2-year follow-up. Methods: A total of 148 patients with idiopathic clubfoot treated surgically by a single surgeon over 15 years were identified, and mailed PODCI questionnaires. Fifty percent of the patients were located and responded, resulting in 74 complete questionnaires. Median age at surgery was 10 months (range, 5.3 to 84.7 mo), male sex 53/74 (71.6%), bilateral surgery 31/74 (41.9%), and average follow-up of 9.7 years. PODCI responses were compared with previously published normal healthy controls using t test for each separate category. Included in the methods is the individual surgeon’s operative technique. Results: In PODCIs where a parent reports for their child or adolescent, there was no difference between our data and the healthy controls in any of the 5 categories. In PODCI where an adolescent self-reports, there was no difference in 4 of 5 categories; significant difference was only found between our data (mean = 95.2; SD = 7.427) and normal controls (mean = 86.3; SD = 12.5) in Happiness Scale (P = 0.0031). Discussion: In this group of idiopathic clubfoot patients, treated with judicious posteromedial release by a single surgeon, primarily when surgery was treatment of choice for clubfoot, patient-based outcomes are not different from their normal healthy peers through childhood and adolescence. While Ponseti treatment has since become the treatment of choice for clubfoot, surgical treatment, in some hands, has led to satisfactory results. Level of Evidence: Level III. Key Words: idiopathic clubfoot, talipes equinovarus, clubfoot treatment
From the *Department of Orthopaedics, Boston Children’s Hospital; and wDepartment of Orthopaedics, Harvard Medical School, Boston, MA. The authors declare no conflicts of interest. Reprints: Susan T. Mahan, MD, MPH, Department of Orthopaedics, Harvard Medical School, 300 Longwood Ave, Boston, MA 02115. E-mail: [email protected]. Copyright r 2014 by Lippincott Williams & Wilkins
J Pediatr Orthop
Volume 34, Number 6, September 2014
(J Pediatr Orthop 2014;34:631–638)
I
n the past decade, treatment of idiopathic clubfoot has shifted from a standard of surgical management to the Ponseti technique.1 Long-term results of the Ponseti technique have been remarkable, which are principally credited to Ponseti himself.2 Long-term surgical outcomes, however, have varied with most authors reporting discouraging outcomes3–6 and some reporting favorable results.7 The majority of comparative studies, albeit retrospective and with shorter follow-up, have favored the Ponseti technique.5,6,8,9 Although Ponseti treatment remains the standard, and has been found to have shortterm success in many different hands,10,11 surgical treatment outcomes varied depending on the treating surgeon and their technique. Although most outcome reports in surgically treated clubfoot discuss results after the Turco technique,3,12,13 more recent use of a judicious, or a la carte posteromedial and lateral release12–14 may result in improved surgical outcomes. Furthermore, “a la carte” surgery is often discussed, but it is not always clear what is on the “menu” and the order of the various procedures. Assessing surgically treated clubfoot can be also difficult, as the proper comparison should be to normal feet, and the assessor should not be the physician but the patient and/or their guardian. Patient-based outcomes comparing children treated for clubfoot to normal healthy controls may be the best assessment of how these feet are functioning. Our purpose was to evaluate a single surgeon’s series of children with idiopathic clubfoot treated with a specific a la carte posteromedial and lateral release using the Pediatric Outcomes Data Collection Instrument (PODCI) with minimum 2-year follow-up. Our primary hypothesis was that children treated surgically for idiopathic clubfoot would be no different in any of the categories of the PODCI when compared with previously published normal healthy controls.15 We also delineate the surgeon’s individual technique.
METHODS After Institutional Review Board approval, we retrospectively identified 148 patients treated surgically by a www.pedorthopaedics.com |
631
Mahan et al
single surgeon (J.R.K.) for idiopathic clubfoot over 15 years, from 1991 through 2006. This was during the time when surgery was the principal treatment option for clubfoot. All 148 patients were diagnosed with idiopathic clubfoot, defined by talipes equinovarus (ICD-9: 754.51) or congenital clubfoot (ICD-9: 754.70) not associated with arthrogryposis, spina bifida, or other conditions known to influence the treatment of clubfoot. All patients’ initial treatment consisted of serial long leg casting but not the Ponseti sequence of casts. Initial clubfoot severity score was not recorded. Tenotomy and foot abduction bracing was not utilized in this group. This treatment was sufficient in approximately 20% of patients and they did not require further surgery. The remaining majority of patients, however, went on to a surgical release, either medial and posterior, or posterior only. See the surgical technique noted below. Patients’ charts were retrospectively reviewed, including operative reports and clinic notes. Patient demographics were noted, including: date of surgery, age at surgery, age at time of analysis, bilaterally of the clubfoot, sex, any complication of the surgery, type of surgery, need for repeat surgery, and recurrence of the clubfoot. Each patient and/or parent was mailed a cover letter, consent form, and an age-appropriate version of the PODCI. An option to complete the instrument online was also provided. If the patient was scheduled for clinic follow-up, the PODCI was made available to them for completion at that time. One hundred forty-eight patients treated surgically for idiopathic clubfoot were mailed a questionnaire. Thirty-one (21%) were returned to sender with no forwarding address. One patient returned the questionnaire indicating that they did not wish to participate. Two patients returned an incomplete questionnaire. Forty patients (27%) did not respond to the questionnaire despite repeat mailings and telephone calls. Seventy-four replied with complete questionnaires, such that 50% of patients were located and included. Patients completed the questionnaire by mail, online, or in clinic (Fig. 1). In the 74 patients who responded to the survey, the median age at surgery was 10 months (range, 5.3 to 84.7 mo), male sex 53/74 (71.6%), bilateral surgery 43/74 (58.1%), and average follow-up at time of survey completion was 9.7 years (minimum 2.0 y, maximum 17.2 y, SD = 4.1 y). In addition, 17.6% (13/74) of the children needed further surgery performed after the initial a la carte posterior only or medial and posterior release. A comparison of patients who did not respond to the survey to those who did respond to the survey was completed to determine if there was any systemic difference between the 2 groups. Initial severity score was not recorded and therefore could not be compared (Table 1). The patients who did not respond to the survey were significantly further out from the surgery temporally (P < 0.005) compared with the responders. There were no other differences between responders and nonresponders in age at surgery, unilateral versus bilateral, sex, or type of surgery (medial and posterior or posterior only) performed.
632 | www.pedorthopaedics.com
J Pediatr Orthop
Volume 34, Number 6, September 2014
Five of the PODCI16 scales were utilized: (1) Transfers and Basic Mobility Scale; (2) Sports/Physical Function Scale; (3) Pain/Comfort Scale; (4) Happiness Scale; and (5) Global Functioning Scale. Patients were divided into the 2 groups as necessary for the PODCI assessment: pediatric (age 2 to 10 y old) and adolescent (age 11 to 18 y old). For the adolescent group, we sent 2 questionnaires: the adolescent self-reported, and the adolescent parent-reported questionnaire. For the pediatric group, the questionnaire was designed to be completed by the parent or guardian. Responses were calculated by the scoring algorithms as created for version 2.0 of the AAOS/COMSS/COSS Outcomes Data Collection Instruments. The standardized score for the PODCI was utilized; all scores are scaled as directed by the PODCI development group from 0 to 100 with the higher score indicating better function or happiness. PODCI responses were compared with previously published normal healthy controls15 using t test for each separate category. We had 39 responses for parent reporting on a pediatric questionnaire (our comparative group had n = 29).15 For adolescent parent-reported PODCI we had 34 responses except for the global functioning score where, because of an incomplete questionnaire, we had only 33 responses (our comparative group had n = 28). For the adolescent self-reported, we had 24 responses (the comparative group had n = 27). We also assessed the responders whose scores fell >2 SD outside the means reported by Haynes et al15 to try and help determine the functional limitations and clinical outcome of these patients. This was a descriptive analysis using their clinic notes and medical records. Additional comparisons were done between patients who had a posterior only versus medial and posterior release. Finally, an assessment of the athletic involvement of these patients was done using question 52 of the PODCI: “Can you/your child participate in competitive level sports with other children the same age (eg, hockey, basketball, soccer, football, baseball, swimming, running [track or cross country], gymnastics or dance)?”
Data Analysis Age at surgery was noted in months (continuous variable). Dichotomous variables, such as sex, bilateral versus unilateral surgery, and need for repeat surgery was noted. Length of follow-up between the surgery and the PODCI assessment was recorded in months (continuous variable). Type of surgery was determined by review of the operative notes, and was a dichotomous variable (posterior and medial release vs. posterior only release). The questionnaires were divided into 3 groups: (1) parent reporting on a child aged 2 to 10 years; (2) parent reporting on an adolescent aged 11 to 18 years; and (3) adolescent aged 11 to 18 years self-reporting. Within each group and scale, a mean score, SD, and sample size was calculated using SPSS (SPSS for Windows, Rel.15.0.0.2006, SPSS Inc., Chicago). These were then compared with student t test assuming different SDs and 2-tailed P-value to previously published normal healthy r
2014 Lippincott Williams & Wilkins
J Pediatr Orthop
Volume 34, Number 6, September 2014
Surgical Treatment of Clubfoot
Contacted patients by mail or clinic
148
Participant
Non-participant
74
74
Respond to mailing
Online
Chose not to participate or didn't complete survey
Return to sender/ no address found
48
5
3
31
Filled out in clinic
No reponse to mailing
21
40
FIGURE 1. Analysis of patients included and not included in the study.
controls15 of similar age, questionnaire group, and scale. This was done using GraphPad version 3.1 (InStat, San Diego, CA). The 2-tailed, unpaired t test, Welch corrected was utilized. It was assumed Gaussian distributions but with populations that may have different SDs. Statistical significance was set a priori at P < 0.05.
Surgical Technique by James R. Kasser, MD The surgical procedure to correct persistent equinovarus deformity of the foot which I used from 1987 until 1998 was rather consistent. Over the first 5 years of practice, to approximately 1987, the procedure evolved becoming more limited. During the evolution, I concentrated the release increasingly on the posterolateral side of the hindfoot, tightened the repair of the heel cord, increased the release of the calcaneocuboid joint from the
medial side, and maintained the spring ligament of the talonavicular joint if possible. Patients had either a posterior only release or a medial and posterior release, depending on the needs of the individual foot. All of the patients in which there was a medial release of the talonavicular joint were pinned antegrade inserting the pin from the posterior aspect of the talus through the head of the talus, centrally through the navicular and exiting between the first and second ray. I did not take x-rays in the OR but confirmed the position of the pin by direct inspection and palpation to be sure that the navicular was neither subluxed dorsally nor laterally. In many cases this meant that there was still some navicular prominence medially after the talonavicular joint was corrected but dorsal lateral subluxation was avoided. Plantar fascia releases were carried out only if there was persistent
TABLE 1. Responders Compared With Nonresponders Analysis Responders N = 74
Nonresponders N = 74
Median age at surgery (mo)
9.63
9.54
Surgery* [n (%)] Unilateral Bilateral Average time from surgery to survey (y)
43 (58) 31 (42) 9.7
38 (58) 27 (42) 11.9
Male sex [n (%)]
54 (73)
53 (71.6)
Type of surgery performed [n/N (%)] Medial and posterior Posterior only
53/74 (72) 21/74 (28)
42/74 (57) 32/74 (43)
Test and P Mann-Whitney U test P > 0.05 Fisher exact P > 0.05 t test P = 0.003 Fisher exact P > 0.05 Fisher exact P > 0.05
*There were 9 nonresponders who had unknown surgical laterality due to unavailability of records.
r
2014 Lippincott Williams & Wilkins
www.pedorthopaedics.com |
633
Mahan et al
flexion deformity through the midfoot. There were no tendon transfers done at the time of initial surgery. The details of the surgical procedure are as follows: The patient was placed in a supine position with a roll under the opposite pelvis. A Cincinnati incision was made, carrying the dissection from the talonavicular joint on the medial side of the foot extending just above the heel crease posteriorly reaching a point beyond the fibula on the lateral side of the foot. The dissection was then carried down through subcutaneous tissue, elevating flaps dorsal and plantar. Sural nerve and lesser saphenous vein were preserved in this dissection. I first entered the peroneal sheath laterally and isolated the peroneal tendons above the level of the distal fibula. I then isolated the heel cord posteriorly and raised a dorsal flap to at least 4 cm. Then the dissection was carried medially, the fascia over the neurovascular bundle was preserved and dissected over the medial malleolus onto the dorsum of the foot. I ensured that the dorsal flap was elevated beyond the midportion of the dorsum of the foot, to allow safe dorsal release of the talonavicular joint. The neurovascular bundle and tendons were preserved in the dorsal layer. The anterior tibial tendon was identified and the dissection carried distally to the most proximal point of its insertion at the first cuneiform. The neurovascular bundle was isolated at the medial malleolus with the dissection carried proximally and distally, defining the medial plantar nerve, the neurovascular bundle, and the calcaneal branch of the nerve. The tendon sheaths of the flexor digitorum longus and posterior tibia were entered with a Z-lengthening of the posterior tibial tendon. In moderate feet, pulleys of tendon sheath were left at the medial malleolus, whereas in severe feet, the tendon sheath was entirely opened across the medial malleolus. In mild feet, the flexor digitorum longus tendon was not lengthened. The posterior tibial was Z-lengthened in all feet. I then opened the talonavicular joint, carrying the dissection first medially, then dorsally, and finally plantarward to the extent necessary to rotate and slide the navicular around the head of the talus. If there was medial subluxation of the cuboid on the calcaneous in severe feet, the medial aspect of the calcaneocuboid joint was entered through the talonavicular joint. The lateral approach to the calcaneocuboid joint and resection of a portion of the cuboid or calcaneus was rarely, if ever, done and only in severe feet. Next, the heel cord was lengthened in a Z-fashion over a distance of 3 to 4 cm. The patient then had the flexor hallucis longus tendon isolated, the tendon sheath opened but the tendon was not lengthened at this point. The extent of opening the tendon sheath was to the posterior aspect of the subtalar joint. One tine of the scissors was then placed through the floor of the tendon sheath of the flexor hallucis longus to open the subtalar joint. This opened the posterior subtalar joint and posterior tibial talar (ankle) joint. I divided the capsule of the subtalar joint to the calcaneofibular ligament. The ankle joint was opened from the flexor hallicus tendon sheath around laterally to the fibula. At this point, generally feet
634 | www.pedorthopaedics.com
J Pediatr Orthop
Volume 34, Number 6, September 2014
will dorsiflex above neutral nicely and that was the extent of the dissection. In occasional cases, more extensive medial capsule was opened. In no cases was the deltoid ligament taken down; in no cases was the interosseous ligament divided. The talonavicular joint was pinned as stated. In severe feet, the tendons of the flexor hallicus and flexor digitorum longus tendon were sewn together proximally and distally and the combined tendon lengthened in a Z-fashion. In mild feet, either the flexor tendons of the toes were not lengthened or lengthened in a fractional way. The Z-lengthening was occasionally done in flexor tendons to the toes but only in older children. The wound was closed as much as possible; the posterior skin was left open if necessary and many feet had a gap as wide as 1.5 cm posteriorly. The neurovascular bundle was covered with soft tissue; a bivavled cast was placed initially in equinus with the foot dorsiflexed under anesthesia at 1 week postoperatively. Again, if the wound remained open, it was simply covered with a nonadherent dressing. The long leg cast remained in place for 2 months; the pin was removed at 1 month. The patient was then started on exercises and occasional postoperative bracing or casting was needed, but in general, simply physical therapy was carried out with no long-term bracing. Many patients did not require the medial release of the talonavicular joint and had a posterior release only. This included Z-lengthening of the Achilles, capsular release of the subtalar and ankle joints, and release of the calcaneofibular ligament. In these patients with generally less involved clubfeet, no pin was placed.
RESULTS There were no statistical differences found among the 5 scales in the pediatric PODCI (aged 2 to 10 y) and the adolescent parent-reported PODCI (aged 11 to 18 y) between our clubfoot patients and normal healthy historical controls. Analysis of the, adolescent self-reported PODCI (aged 11 to 18 y) showed one statistically significant difference in the Happiness Scale, in which our surgically treated clubfoot patients were happier than the normal healthy historical controls (our mean 95.2 and SD = 7.427 compared with historical controls mean 86.3 and SD = 12.5; P = 0.003) (Table 2). In the analysis of the responders who were outside 2 SD of the means reported by Haynes et al,15 there were 22 patients (31.9%) who fell
