Slipped capital femoral epiphysis (SCFE) is one of the most common conditions treated surgically by pediatric orthopedists. With a national incidence of approximately 10 adolescents per 100,000, SCFE is most frequently seen in boys, with a sex ratio of 1.7 male to female.1 Incidence has been shown to correlate with geographic location and race, with racial minorities in the United States more commonly affected.2–5
Slipped capital femoral epiphysis occurs when the femoral head metaphysis is displaced anterosuperiorly in relation to the physis.6 Pathogenesis is multifocal and based on forces of stress along the physis or reduced ability to handle shear strain along the physis.7–9 Many diseases have been implicated in SCFE, including obesity beyond the 95th percentile for body mass index, endocrinopathies, and systemic disorders.10–15 Slipped capital femoral epiphysis bilaterality in patients with endocrinopathies, such as hypothyroidism, has been observed in up to 61% of patients.12
Although controversy exists regarding optimal treatment, historic principles include preventing further slip progression and deformity by stabilizing the proximal femoral epiphysis. New procedures, such as the modified Dunn procedure for capital realignment, have gained favor by providing the surgeon the opportunity to restore anatomic alignment while visualizing the vulnerable femoral head vasculature. Regardless of procedure, there are numerous well-documented complications including slip progression, hardware failure, contralateral slip, chondrolysis, and avascular necrosis.
Overall, little data exist regarding nationally representative SCFE complication and readmission rates. The authors hypothesized that cardinal differences exist for SCFE readmission based on patient's characteristics and surgeon's choice of procedure.
Materials and Methods
The Healthcare Cost and Utilization Project's (HCUP) Nationwide Readmissions Database (NRD) was created by the Agency for Healthcare Research and Quality to address the gap of nationally representative health care readmission data. This database was first released for the 2013 year and is derived from the HCUP State Inpatient Databases (SID). The large sample allows researchers to study rare and uncommon disorders and procedures for patients of all ages with a nationally representative scale. This discharge level database in 2013 contains 36 million weighted discharges from 21 states and accounts for 49.1% of all US hospitalizations.16 The NRD is post stratified and weighted, with all values given on a nationally representative scale for patients discharged from community hospitals with the exclusion of noncommunity hospitals, rehabilitation hospitals, or long-term acute care facilities. This database provides data on the number of times a unique patient was readmitted within that year, the diagnoses and procedures done at each visit, patient demographics and clinical characteristics, comorbidities, and hospital characteristics.
The 2013 NRD was queried using International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes to identify patients with a primary diagnosis of SCFE (ICD-9-CM code 732.2) between 9 and 16 years old.
Readmission was evaluated for any cause within 90 days of the discharge date. Days between admissions were defined as the time spent between the patient discharge date and the readmission date. The discharge date was defined as the admission date plus patient length of stay (LOS). If LOS was missing, the current authors were not able to calculate days between admissions and the case was excluded.
During the 90-day all-cause readmission analysis, the months of January to September were analyzed to ensure that if a patient was admitted on the last day of September, the data could capture a potential readmission in December. The NRD cannot capture readmission data if it falls into the subsequent year, so the last 90 days of the year were not included in the analysis to create a true 90-day readmission rate, as recommended by the HCUP.16 For a primary diagnosis of SCFE, the current authors analyzed patients' demographics and clinical characteristics, initial procedure types, readmission rates, mean days until readmission, diagnosis on readmission, and procedure on readmission.
Three procedure-specific cohorts were created to compare the choice of treatment and associated outcome. The first category was for patients with a closed procedure who had ICD-9-CM code 79.45 (closed reduction of separated epiphysis of femur) or 78.55 (internal fixation of femur). A second category was created for patients with open procedures when they had ICD-9-CM codes 79.55 (open reduction of separated epiphysis of femur) or 79.35 (open reduction with internal fixation of femur). Finally, a third category was created for patients who had both open and closed procedures on the same visit. Subsequent analyses include distinguishing individual ICD-9-CM procedure types for readmission rates. Patient demographics and clinical characteristics provided by the NRD used in the study included age, sex, race, and LOS. A previously established index for calculating comorbidities using the ICD-9-CM codes was used to find patients who had comorbidities, including obesity, as well as metabolic, systemic, nervous, renal, cardiopulmonary, and other multi-system diseases and disorders.17,18
Univariate and multivariate techniques were used to create models based on binary logistic regression and linear regression. Continuous variables are tested for linearity using the Box-Tidwell procedure to ensure the model variable is linearly related to the logit of the dependent variable. Bonferroni correction was used when multiple independent variables were compared to circumvent the associated risk of type I error.
The P values were reported as 2-sided and additional statistical tests, such as chi-square test or Fisher's exact test, were used when appropriate. Computations were done using SPSS Statistics version 23.0 (SPSS Inc, Chicago, Illinois) and SAS version 9.4 (SAS Institute, Cary, North Carolina) software. To protect patient confidentiality, the HCUP NRD data user agreement prohibits the publication of patient discharges when there are 10 or fewer cases present. Therefore, “≤10” was used when appropriate or “0” was used when there were no occurrences of the data.
A total of 1415 patients 9 to 16 years old with a primary diagnosis of SCFE were identified in the 2013 NRD. For the months January to September, there were 1082 patients with a primary diagnosis of SCFE. A total of 158 patients would be readmitted during these months; however, only 58 (5.9%) patients had a readmission date within 90 days. Extending analysis to readmissions beyond 90 days would require an equally extended window at the end of the year to make the data statically valid. Therefore, these 100 patient readmissions were outside of the 90-day window and were not analyzed to maintain statistical accuracy. Of the remaining 982 patients, 58 (5.9%) were readmitted within 90 days of the index procedure (Table 1).
Univariate Analysis of 90-Day Readmission Rates for Slipped Capital Femoral Epiphysis Procedures Among Patients 9 to 16 Years Old in 2013 (January to September)
A total of 116 procedures were done for the 58 patients who would later be readmitted, and 64 of those SCFE procedures were open, closed, or both. The remaining 52 procedures were not clinically meaningful or statistically significant (Table 2). Having a bilateral closed procedure was not a predictor of read-mission (P=.583) and there were no patients with bilateral open procedures that would subsequently be re-admitted. Equal percentages of bilateral closed procedures were found between the readmitted (27.4%) and not readmitted (27.4%) groups. However, having both an open and closed procedure accounted for a greater portion of the readmitted (22.7%) vs not readmitted (2.9%) group (P<.001).
All Cause 90-Day Admission and Readmission for Slipped Capital Femoral Epiphysis Among Patients 9 to 16 Years Old in 2013 (January–September)
Patients readmitted were on average 6 months younger (mean, 11.5 years) than those not readmitted (mean, 12.0 years) (P=.029). The odds of readmission decreased with increasing age (odds ratio [OR], 0.830). Male sex increased the odds of readmission 1.830 times (OR, 1.830; 95% confidence interval [CI], 1.009–3.319; P=.047). Median household income was not found to be a predictor of readmission rates (P=.887) (Table 1).
Of the 58 patients who were readmitted, the most common reason for readmission was nontraumatic SCFE (27%; ICD-9-CM code 732.2). Due to the nature of the database, it is unclear whether this occurred on the same side as the index procedure or if this represented a new contralateral event. Fifteen (25%) patients were readmitted for mechanical complication of orthopedic implant and 13 (22%) for subtrochanteric femur fracture.
On readmission, the most common secondary procedure was an open procedure. Of the patients who underwent a secondary procedure, 22 (37%) had open reduction, 10 (18%) underwent open reduction internal fixation, and 10 had internal fixation of femur. No patients who were readmitted underwent a secondary closed reduction (Table 2).
Comorbidities were present in 46.6% of patients with SCFE the current authors studied and in 79.3% of patients who would later be readmitted. Of patients who were readmitted, 20.7% had 1 comorbidity and 29.3% had 2 comorbidities. For the cohort that was not readmitted, 25.8% had 1 comorbidity, 6.7% had 2 comorbidities, and less than 2% had 2 or more comorbidities.
Complications were observed in those who were readmitted in 5.2% (n≤10) of cases and not readmitted in 4.4% (n=41) of cases. Neither patient cohort had more than 1 complication at discharge and neither cohort had any cases of death. Complications at the time of readmission and prior include surgical site infection and transfusion.
Readmission within 90 days following a procedure for a primary diagnosis of SCFE was analyzed using binary logistic regression with respect to comorbidities, LOS, age, sex, transfusion as a complication, number of diagnoses on record, and the likelihood of being readmitted. Comorbidities observed in both readmitted and nonreadmitted patients were hypothyroidism, obesity, and chronic pulmonary disease. Overall, the model was found to be statistically significant (chi-square, 72.927; P<.001). Statistically significant predictors included increasing LOS with an increased odd of being readmitted by 5.383 times (OR, 5.383; 95% CI, 1.079–26.846; P=.040). Increasing age between 9 and 16 years was a predictor of decreased likelihood of being readmitted (OR, 0.797; 95% CI, 0.641–0.992; P=.042).
The most important predictor in the model was comorbidity of hypothyroidism. When present, a patient was 47.4 times more likely to be readmitted in this model (95% CI, 13.807–163.011; P<.001). When chronic pulmonary disease was considered as an individual comorbidity, it was found to be statistically significant as a predictor of increased likelihood for readmission (OR, 2.748; 95% CI, 1.513–4.992; P=.002).
All patients readmitted with hypothyroidism (n=15) had a primary readmission diagnosis of “Unspecified mechanical complication of internal orthopedic device, implant, and graft” (ICD-9-CM code 996.49). Other diagnoses on re-admission included the following: pediatric body mass index greater than or equal to 95th percentile (ICD-9-CM code V8554), obesity (ICD-9-CM code 278.00), acute posthemorrhagic anemia (ICD-9-CM code 285.1), and unspecified vitamin D deficiency (ICD-9-CM code 268.9). Twelve of those readmitted had procedures for open reduction, and 10 or fewer had a bone graft procedure (ICD-9-CM codes 77.77, 78.05).
This study sought to identify factors associated with SCFE readmission and patients who may be at an increased risk for readmission. No studies to date have described SCFE readmission reasons on a nationally representative scale and no guidelines exist for the optimal choice of treatment in many atypical SCFE cases. The predictive models created in this study can help identify patients at an increased risk for readmission and short-term complications.
Hypothyroidism was strongly associated with readmission (OR, 35.465; P<.001) in the current study, all occurring due to mechanical complications. Physicians should consider endocrine evaluation in patients with SCFE who present with atypical age (younger than 10 years, older than 16 years) or have a height greater than the third percentile.19 In addition, hypothyroidism should lead to consideration of increased mechanical stability, longer duration of protected weight bearing, and contralateral prophylactic fixation.
Obesity as an association with increased incidence of SCFE has been well documented in a multitude of different geographic locations and patient cohorts.15,20,21 The current authors found obesity to be a non-statistically significant predictor for readmission with univariate analysis (P=.669). This is likely due to underreporting of obesity in the database.
Both closed and open reduction as initial procedures were associated with increased risk of readmission (P=.003 and P<.001, respectively). This could indicate an increased likelihood of complications requiring admission or increased risk of contralateral slip when reduction is required with the index procedure. These findings are in agreement with previous studies documenting risks of complications related to closed and open reductions.22–24
There were several important limitations to this study. First, the NRD is a derivative of the SID and inherently depends on the integrity and quality of data within the SID. Choosing a window of analysis is difficult and requires clinical reasoning more so than mathematical reasoning. The 90-day window for readmission was chosen as a good representation of acute and subacute SCFE complications that could be captured. The procedures not captured in the last 90 days of the year did not significantly vary from the procedures in the other months. Although the procedures were nearly identical in ratio, it is possible that factors such as ultraviolet light exposure in the winter months are important as predictors of complications and readmission. Recently, seasonal variations have been implicated as playing an important role in SCFE.25
In this study, 5.9% of patients with SCFE were readmitted, and readmission was more common among younger patients, males, and patients who underwent open or closed reduction. Implant-related complications occurred in 25% of patients on readmission. Thirty-seven percent of patients had an open reduction at read-mission. Hypothyroidism is a significant predictor for readmission within 90 days; however, obesity is not (likely due to underreporting). Additional study is needed to determine how to avoid mechanical failures that are associated with readmissions.
- Lehmann CL, Arons RR, Loder RT, Vitale MG. The epidemiology of slipped capital femoral epiphysis: an update. J Pediatr Orthop. 2006;26(3):286–290. doi:10.1097/01.bpo.0000217718.10728.70 [CrossRef]16670536
- Benson EC, Miller M, Bosch P, Szalay EA. A new look at the incidence of slipped capital femoral epiphysis in New Mexico. J Pediatr Orthop. 2008;28(5):529–533. doi:10.1097/BPO.0b013e31817e240d [CrossRef]18580367
- Phadnis J, Phillips P, Willoughby R. The epidemiologic characteristics of slipped capital femoral epiphysis in Maori children. J Pediatr Orthop. 2012;32(5):510–514. doi:10.1097/BPO.0b013e31824b2b4c [CrossRef]22706468
- Loder RT, Skopelja EN. The epidemiology and demographics of slipped capital femoral epiphysis. ISRN Orthop. 2011;2011:486512. doi:10.5402/2011/486512 [CrossRef]24977061
- Nguyen AR, Ling J, Gomes B, Antoniou G, Sutherland LM, Cundy PJ. Slipped capital femoral epiphysis: rising rates with obesity and aboriginality in South Australia. J Bone Joint Surg Br. 2011;93(10):1416–1423. doi:10.1302/0301-620X.93B10.26852 [CrossRef]21969445
- Aronsson DD, Loder RT, Breur GJ, Weinstein SL. Slipped capital femoral epiphysis: current concepts. J Am Acad Orthop Surg. 2006;14(12):666–679. doi:10.5435/00124635-200611000-00010 [CrossRef]17077339
- Fishkin Z, Armstrong DG, Shah H, Patra A, Mihalko WM. Proximal femoral physis shear in slipped capital femoral epiphysis: a finite element study. J Pediatr Orthop. 2006;26(3):291–294. doi:10.1097/01.bpo.0000217730.39288.09 [CrossRef]16670537
- Litchman HM, Duffy J. Slipped capital femoral epiphysis: factors affecting shear forces on the epiphyseal plate. J Pediatr Orthop. 1984;4(6):745–748. doi:10.1097/01241398-198411000-00017 [CrossRef]6511903
- Pritchett JW, Perdue KD. Mechanical factors in slipped capital femoral epiphysis. J Pediatr Orthop. 1988;8(4):385–388. doi:10.1097/01241398-198807000-00001 [CrossRef]3392189
- Witbreuk M, van Kemenade FJ, van der Sluijs JA, Jansma EP, Rotteveel J, van Royen BJ. Slipped capital femoral epiphysis and its association with endocrine, metabolic and chronic diseases: a systematic review of the literature. J Child Orthop. 2013;7(3):213–223. doi:10.1007/s11832-013-0493-8 [CrossRef]
- Aversano MW, Moazzaz P, Scaduto AA, Otsuka NY. Association between body mass index-for-age and slipped capital femoral epiphysis: the long-term risk for subsequent slip in patients followed until physeal closure. J Child Orthop. 2016;10(3):209–213. doi:10.1007/s11832-016-0731-y [CrossRef]27095178
- Loder RT, Wittenberg B, DeSilva G. Slipped capital femoral epiphysis associated with endocrine disorders. J Pediatr Orthop. 1995;15(3):349–356. doi:10.1097/01241398-199505000-00018 [CrossRef]7790494
- Bhatia NN, Pirpiris M, Otsuka NY. Body mass index in patients with slipped capital femoral epiphysis. J Pediatr Orthop. 2006;26(2):197–199. doi:10.1097/01.bpo.0000218526.36362.3f [CrossRef]16557134
- Dunbar J, Goulding A. Slipped capital femoral epiphysis: more New Zealand cases likely as obesity rises in children and adolescents?N Z Med J. 2001;114(1145):559–560.
- Murray AW, Wilson NI. Changing incidence of slipped capital femoral epiphysis: a relationship with obesity?J Bone Joint Surg Br. 2008;90(1):92–94. doi:10.1302/0301-620X.90B1.19502 [CrossRef]
- Agency for Healthcare Research and Quality. Introduction to the HCUP nationwide readmissions database (NRD)2013. www.hcup-us.ahrq.gov/db/nation/nrd/Introduction_NRD_2013.pdf. Accessed October 2017.
- Elixhauser A, Steiner C, Harris DR, Coffey RM. Comorbidity measures for use with administrative data. Med Care. 1998;36(1):8–27. doi:10.1097/00005650-199801000-00004 [CrossRef]9431328
- Menendez ME, Ring D. A comparison of the Charlson and Elixhauser comorbidity measures to predict inpatient mortality after proximal humerus fracture. J Orthop Trauma. 2015;29(11):488–493. doi:10.1097/BOT.0000000000000380 [CrossRef]26165266
- Moyer J, Jacks L, Hunter JD, Chan G. Slipped capital femoral epiphysis and associated hypothyroidism: a review of the literature with two classic case examples. J Pediatr Endocrinol Metab. 2016;29(4):427–434. doi:10.1515/jpem-2015-0311 [CrossRef]26812778
- Manoff EM, Banffy MB, Winell JJ. Relationship between body mass index and slipped capital femoral epiphysis. J Pediatr Orthop. 2005;25(6):744–746. doi:10.1097/01.bpo.0000184651.34475.8e [CrossRef]16294129
- Burrows HJ. Slipped upper femoral epiphysis; characteristic of a hundred cases. J Bone Joint Surg Br. 1957;39(4):641–658. doi:10.1302/0301-620X.39B4.641 [CrossRef]13491627
- Kitano T, Nakagawa K, Wada M, Moriyama M. Closed reduction of slipped capital femoral epiphysis: high-risk factor for avascular necrosis. J Pediatr Orthop B. 2015;24(4):281–285. doi:10.1097/BPB.0000000000000170 [CrossRef]25812031
- Canale ST. Problems and complications of slipped capital femoral epiphysis. Instr Course Lect. 1989;38:281–290.2649577
- Palocaren T, Holmes L, Rogers K, Kumar SJ. Outcome of in situ pinning in patients with unstable slipped capital femoral epiphysis: assessment of risk factors associated with avascular necrosis. J Pediatr Orthop. 2010;30(1):31–36. doi:10.1097/BPO.0b013e3181c537b0 [CrossRef]
- Loder RT, Schneble CA. Seasonal variation in slipped capital femoral epiphysis: new findings using a national children's hospital database. J Pediatr Orthop. 2019;39(1):e44–e49. doi:10.1097/BPO.0000000000001074 [CrossRef]
Univariate Analysis of 90-Day Readmission Rates for Slipped Capital Femoral Epiphysis Procedures Among Patients 9 to 16 Years Old in 2013 (January to September)a
|Parameter||Readmitted (n=58)||Not Readmitted (n=924)||Odds Ratio||95% Confidence Interval||P|
|National Count||Percent||National Count||Percent|
|Primary diagnosis of slipped capital femoral epiphysis (732.2), No.||58||5.9||924||94.1|
|Age, mean±SD, y||11.5±1.4||12.0±1.7||0.830||0.702–0.981||.029|
|Total length of stay, mean±SD (range), d||2.5±1.3 (1–5)||2.2±2.9 (0–36)||1.029||0.955–1.108||.456|
|Closed or open procedures, No.b,c||64||857|
|Closed procedures, No.b||47||73.4||717||83.6||1.773||1.008–3.120||.047|
|Open procedures, No.c||≤10||#||113||13.2||2.964||0.823–10.674||.096|
|Both open and closed procedures, No.d||15||22.7||27||2.9||0.090||0.044–0.181||<.001|
|Diagnoses on discharge, mean±SD (range), No.||3.0±2.0 (1–6)||2.3±1.8 (1–14)||1.155||1.035–1.290||.010|
|Procedures on discharge, mean±SD (range), No.||1.8±1.1 (0–5)||1.6±0.9 (0–5)||1.245||0.959–1.618||.100|
|E-codes on discharge, mean±SD (range), No.||0.04±0.2 (0–1)||0.1±0.42 (0–4)||0.743||0.299–1.845||.522|
|Chronic conditions on discharge, mean±SD (range), No.||2.0±1.2 (1–6)||1.6±1.0 (1–8)||1.323||1.076–1.626||.008|
|Median household income quartile, No.||.887|
| $64,000 or more||12||176||0.992||0.450–2.190||.985|
|Patient state same as hospital state, No.||58||100.0||878||95.0||0.000||0.000–0.000||.998|
|Weekend admission, No.||≤10||#||64||6.9||0.863||0.322–0.2308||.769|
All Cause 90-Day Admission and Readmission for Slipped Capital Femoral Epiphysis Among Patients 9 to 16 Years Old in 2013 (January–September)a
|Parameter||Readmitted (n=58)||Not Readmitted (n=924)||Odds Ratiob||95% Confidence Interval||P|
|National Count||Percent||National Count||Percent|
| 78.55—Internal fixation of bone without fracture reduction, femur||39||68.0||683||73.9||1.322||0.770–2.271||.312|
| 79.45—Closed reduction of separated epiphysis, femur||23||39.0||178||19.3||0.421||0.236–0.750||.003|
| 79.35—Open reduction of fracture with internal fixation, femur||0||0||15||1.7||-||-||-|
| 79.55—Open reduction of separated epiphysis, femur||17||30.0||92||9.9||0.359||0.198–0.652||<.001|
| 78.25—Limb shortening procedures, femur||0||0||32||3.5||-||-||-|
| 78.45—Other repair or plastic operations on bone, femur||≤10||#||29||3.2||0.723||0.193–2.701||.629|
| 78.65—Removal of implanted devices from bone, femur||≤10||#||34||3.7||0.758||0.217–2.647||.664|
| 38.99—Venous puncture not elsewhere classified||0||0||16||1.7||-||-||-|
| 93.99—Other respiratory procedures||0||0||12||1.4||-||-||-|
| 93.94—Respiratory medication administered by nebulizer||0||0||12||1.4||-||-||-|
| 77.35—Femoral division not elsewhere classified||≤10||#||55||5.9||1.247||0.363–4.283||.726|
| 77.25—Femoral wedge osteotomy||0||0||23||2.5||-||-||-|
| 732.2—Nontraumatic slipped capital femoral epiphysis||16||27.0|
| 996.49—Other mechanical complication of other internal orthopedic device, implant, and/or graft||15||25.0|
| 820.22—Closed fracture of subtrochanteric section of neck of femur||13||22.0|
| 835.03—Other closed anterior dislocation of hipc||≤10||#|
| 732.1—Juvenile osteochondrosis of hip and pelvisc||≤10||#|
| 733.42—Aseptic necrosis of head and neck of femurc||≤10||#|
| 996.40—Unspecified mechanical complication of internal orthopedic device, implant, and graftc||≤10||#|
| 998.59—Other postoperative infectionc||≤10||#|
| 711.05—Pyogenic arthritis, pelvic region and thighc||≤10||#|
| 79.55—Open reduction of separated epiphysis, femur||22||37.0|
| 93.46—Limb skin traction not elsewhere classified||≤10||#|
| 79.35—Open reduction of fracture with internal fixation, femur||≤10||#|
| 78.55—Internal fixation of bone without fracture reduction, femur||≤10||#|
| 78.65—Removal of implanted devices from bone, femur||12||20.0|