Translate this page into:
Impact of delayed presentation and surgical management on radiologic and clinical outcomes of pediatric septic hip
, Asal Seifaeid, Houyar Zarifkard, Amirhossein Kamaliniaa
⁎Corresponding author: Mohammad Hadi Gerami. dr.hadi.gerami@gmail.com
-
Received: ,
Accepted: ,
This article was originally published by Reed Elsevier India Pvt. Ltd. and was migrated to Scientific Scholar after the change of Publisher.
Abstract
Abstract
Septic arthritis of the hip joint (septic hip) is the bacterial infection of the hip. Delayed treatment increases the risk of extensive joint damage, systemic infection, prolonged recovery, higher healthcare costs, and long-term disability. This study explores the repercussions of delayed surgical treatment in pediatric septic hip.
In this observational study, pediatric patients diagnosed with septic hip between 2012 and 2021 were retrospectively selected from a major referral center in Shiraz, Iran. We collected clinical and radiological data, including Kocher score and determined Choi classification in follow-up radiographs. Multivariate logistic regression analysis was used to assess the impact of delay to surgery on the development of clinical sequelae and radiological deformities in the presence of potential confounders of age and Kocher criteria.
Out of 49 children with delayed presentation, 46 survived and entered the study. Mean delay from symptom presentation to admission was 12.67 ± 10.51 days, and mean delay from admission to surgery was 5.33 ± 6.47 days. Of the 46 patients, 28.26% developed clinical sequelae. The Choi classification revealed that 54.35% of patients developed no residual deformity. A multivariate logistic regression analysis indicated a statistically significant association between delay to surgery and the development of radiological deformities (adjusted odds ratio: 1.36, 95% CI: 1.14–1.64, p = 0.001). Additionally, a separate analysis revealed that each additional day of delay was associated with a 12% increase in the odds of clinical sequelae (adjusted odds ratio: 1.12, 95% CI: 1.03–1.22, p = 0.006). Sensitivity analyses confirmed the relationship of delay from admission to surgery in developing both outcomes.
Delay in medical care and surgical management remains the most important factor affecting the outcomes of septic hip. The study underscores the critical role of timely surgical intervention in reducing complications in pediatric septic hip patients.
Keywords
Septic arthritis
Hip joint
Septic hip
Pediatric
Orthopedics
Outcome
1 Introduction
The hip joint is one of the most common sites for infection in the pediatric population, comprising 22.5–43% of all septic arthritis cases.1,2 Adverse outcomes are expected in the late management of this entity, making it an orthopedic emergency. The typical presentation of a septic hip is hip pain, inability to bear weight and fever. However, the challenge remains in differentiation of a septic hip from transient synovitis of hip. The diagnosis is made by a combination of clinical examination and laboratory evaluation.3,4 In 1999, Kocher et al. suggested four main factors differentiating hip septic arthritis. The classic factors were fever ≥38.5 °C, inability to bear weight, serum white blood cells (WBC) ≥12,000/mm³ count, and erythrocyte sedimentation rate (ESR) ≥40 mm/h.5 Since then, modifications have been suggested to be included as a diagnostic factor such as C-reactive protein (CRP) ≥20 mg/L.6,7
Management of hip septic arthritis consists of a combination of antibiotic therapy and operative treatment.3 There is a consensus regarding the necessity of surgical treatment especially in hip septic arthritis. Almost all age groups and microorganisms require urgent surgical irrigation and debridement.8,9
The long-term complications of septic arthritis of the hip in pediatrics can include joint stiffness, avascular necrosis, growth disturbances, limb length discrepancy, and chronic arthritis.
Choi et al. proposed a widely accepted classification system for radiological outcomes of septic hip in pediatric patients in 2005. Type I comprises cases with no deformity (1a) or mild coxa magna (1b). Type II involves the epiphysis, physis, and metaphysis, with presentations of coxa brevis (2a) or progressive coxa vara or valga (2b). Type IIIa is characterized by coxa vara or valga accompanied by femoral neck slipping, while type IIIb involves pseudoarthrosis. Type IV is assigned when there is destruction of the femoral head and neck, with subtypes 4a and 4b depending on the presence of a small remaining neck or complete loss of the head and neck.10
There have been a few studies with limited sample sizes that evaluated septic hip in the pediatric population. Lunseth and Heiple11 reviewed 38 children (39 hips) with septic arthritis to identify key factors influencing hip joint prognosis. Younger patients, especially those under one year of age, showed poorer results in roentgenographic classification three or more years later. Prolonged duration from clinical onset to therapy initiation also correlated with worse outcomes. Causative organisms, open versus closed treatment, and the prevalence of Staphylococcus aureus were explored, revealing no clear correlation with prognosis, though nonstaphylococcal organisms may be less destructive to the femoral head in infants. Frolin and Milani12 investigated the sequelae of the septic hip in 41 hips. In a retrospective study among 31 patients under 18 months old, univariate analysis revealed poorer radiological prognoses in patients with underlying diseases, longer symptom duration, and pus drainage, but only symptom duration was statistically significant in multivariate analysis, underscoring the importance of timely surgical intervention.13 They proposed a radiologic classification that was simpler than Choi's that showed improved concordance to “satisfactory results” which included stability, a range of motion of 50° or more of flexion, a flexion contraction below 20°, and a painless hip. A 1983 study14 also investigated delayed management of septic hip, revealing that prognosis is poorer in younger children and cases with associated osteomyelitis of the proximal femur. Results showed that reconstructive surgery, particularly in neonates, is challenging and may not always yield the desired functional and anatomical improvements. The study emphasized that the time elapsed between symptom onset and treatment is the key factor influencing the ultimate outcome.
In this study, we intend to evaluate comorbidities and complications of septic arthritis of the hip joint, in the pediatric population with delayed surgical management in long-term follow-ups.
2 Methods
This study was written in compliance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guideline.
2.1 Study setting and participants
This is an observational study conducted in Shiraz, southern Iran. We used a retrospective cohort methodology with the exposure being delayed surgical treatment. Patients were retrospectively selected from Namazi (Nemazee) Hospital, a university hospital, and a major referral center affiliated with Shiraz University of Medical Sciences. Inclusion criteria were age under 18 years, at least one day delay between admission to surgery or presentation of symptoms to admission, having a final diagnosis of septic hip, and availability of lab data and radiographs. Exclusion criteria were missing data and inconclusive final diagnosis.
2.2 Data collection
Records of patients who were admitted with the final diagnosis of septic hip between 2012 and 2021 were investigated and their clinical data were extracted. Duration of follow-ups varied from 10 months to 11 years and 3 months (mean: 6 years and 7 months). In case of missing follow-up radiographs, patients were contacted via a phone call and invited to our clinic for radiologic and clinical evaluation. Where radiological data were present, patients were contacted to solicit current medical histories and hip joint-related complaints. Patients who indicated no subjective complaints on phone calls were labeled as having no clinical sequelae. The rest were examined in the clinic and their hip-related complications were noted. The conditions that resulted in prolonged admission or readmission like sepsis or osteomyelitis were also considered as clinical sequelae.
We also collected some clinical and laboratory findings from admission such as on-arrival temperature, ability to bear weight, ESR, and WBC count. These parameters constitute the Kocher score – a score originally introduced to calculate the risk of septic arthritis. We used these as a proxy for the severity of the disease on presentation and entered it in the analysis as a confounder for the outcome. Antibiotic treatment was initiated for all patients under 24 h from admission, therefore we did not include a delay in antibiotic treatment in the analysis.
Follow-up radiographs were evaluated by the first author – an orthopedic surgeon, and they were stratified according to Choi classification. The Choi classification uses radiological data to assess residual deformities in children with septic hip, guiding the appropriate treatment decisions.15 This scale categorizes septic arthritis sequelae into four types (Class I to Class IV) and 8 subtypes (A and B for each class) as described in the introduction.
2.3 Statistical analysis
We created two outcome groups, one with or without residual radiographic deformity, and another with or without hip-related clinical sequelae, including pain, limping, and limitation of motion. We considered Choi Class IA as having no deformity and Classes IB to IV as the presence of deformity. For each outcome we ran a logistic regression model in which we included delay to surgery as an independent variable, along with other related data as confounders. Confounding variables were the Kocher score, age, gender, and duration of symptoms before admission (delayed presentation). Analyses were done using Stata Statistical Software: Release 18. College Station, TX.
Clinical and radiological data were also reported as descriptive statistics.
2.4 Sensitivity analysis
To assess the potential impact of reporting and recall bias in the data related to duration of symptoms prior to admission, a sensitivity analysis was conducted. Since these data were reported by patients' parents or companions, there is a possibility of variability in the accuracy and completeness of the information provided. To address these biases, we performed a sensitivity analysis by removing the variable representing the duration of symptoms prior to admission from our multivariate logistic regression models. Instead we used delay from admission to surgery. This allowed us to evaluate the robustness of findings regarding the impact of delay in surgery and assess whether the inclusion of symptoms before admission had a significant influence on the outcomes.
2.5 Ethical considerations
This study was conducted in accordance with the Declaration of Helsinki and was approved by our institutional review board.
3 Results
In total, 49 pediatric patients had been admitted in the study period. Three passed away during the hospital stay whose case details can be seen in Table 1.
| Gender | Age | Symptom to admission (days) | Admission to surgery (days) | Past medical history | WBC | ESR | Weight-bearing | Fever | Complications |
| Female | 17 years | 4 | 11 | Wilson's disease, cirrhosis | 12,900 | 69 | No | Yes | Local abscess, septic shock |
| Female | 5 years and 6 months | 7 | 8 | Nothing | 7600 | 87 | No | Yes | Myositis, osteomyelitis, septic shock |
| Male | 1 month | 5 | 7 | Nothing | 13,800 | 122 | NA | No | Pelvic abscess, myositis, osteomyelitis |
The remaining 46 completed the follow-up and comprised 27 males (58.70%) and 19 females (41.30%). The clinical characteristics of the patients can be seen in Table 2.
| Age, mean ± SD | 5.37 ± 4.88 | Long-term complications | Pain, n (%) | 7 (15.22) | |
| Gender | Male, n (%) | 27 (58.70) | Limited range of motion, n (%) | 2 (4.35) | |
| Female, n (%) | 19 (41.30) | ||||
| WBC>12,000, n (%) | 22 (47.83) | ||||
| ESR>40 mm/h, n (%) | 40 (86.96) | Limping | 3 (6.52) | ||
| Temperature>38.5, n (%) | 24 (52.17) | In-hospital complications | Osteomyelitis, n (%) | 4 (8.7) | |
| Septic shock, n (%) | 1 (2.17) | ||||
| Non-weight bearing, n (%) | 39 (84.78) | Choi classification in follow-up | 1A, n (%) | 25 (54.35) | |
| Kocher score | 0, n (%) | 1 (2.17) | 1B, n (%) | 5 (10.87) | |
| 1, n (%) | 2 (4.35) | 2A, n (%) | 6 (13.04) | ||
| 2, n (%) | 19 (41.30) | 2B, n (%) | 3 (6.52) | ||
| 3, n (%) | 11 (23.91) | 3A, n (%) | 3 (6.52) | ||
| 4, n (%) | 13 (28.26) | 3B, n (%) | 0 (0) | ||
| Symptoms prior to admission (days), mean ± SD | 12.67 ± 10.51 | 4A, n (%) | 2 (4.35) | ||
| Admission to surgery delay (days), mean ± SD | 5.33 ± 6.47 | 4B, n (%) | 2 (4.35) | ||
Clinical sequelae, either short or long-term, were observed in 13 (28.26%) patients, with 7 (15.22%) reporting pain, 2 (4.35%) experiencing limited range of motion, and 3 (6.52%) exhibiting limping. In-hospital (short-term) complications comprised 1 (2.17%) case of septic shock and 4 (8.7%) developing osteomyelitis.
In the follow-up, Choi's classification revealed that 25 patients (54.35%) belonged to Class 1A, i.e., showing no residual deformity. The rest had varying degrees of deformity as seen in Table 2. Fig. 1 shows follow-up radiographs of two patients.
A multivariate logistic regression analysis was conducted to identify factors associated with radiological deformities in the follow-up of pediatric septic hip patients. Notably, age exhibited a weakly significant association, with an odds ratio of 1.38 (95% CI: 1.01–1.88, p = 0.046), suggesting that older patients had an increased likelihood of developing radiologic deformity. However, gender did not show a significant association with development of radiological deformities. Regarding the Kocher score, compared to lower sores, patients with a Kocher score of 3 or 4 did not exhibit statistically significant associations with radiologic deformity. In contrast, a statistically significant association was observed with delay to treatment, where each additional day of delay from symptoms to surgery increased the odds of deformity by 36% (adjusted odds ratio: 1.36, 95% CI: 1.14–1.64, p = 0.001) (Table 3).
| Adjusted odds ratio | Confidence interval | P value | |||
| Lower | Upper | ||||
| Age | 1.38 | 1.01 | 1.88 | 0.046 | |
| Gender | Male | Ref. | – | – | – |
| Female | 0.32 | 0.04 | 2.75 | 0.299 | |
| Kocher scorea | 0, 1, or 2 | Ref. | – | – | – |
| 3 | 7.57 | 0.17 | 328.20 | 0.293 | |
| 4 | 25.76 | 0.72 | 915.96 | 0.074 | |
| Symptoms to surgery (days) | 1.36 | 1.14 | 1.64 | 0.001 | |
Another multivariate logistic regression analysis aimed to identify factors associated with clinical sequelae in follow-up. The results indicated that age, gender, and Kocher score were not significantly associated with clinical sequelae. A statistically significant association was observed with the delay from admission to surgery, where each additional day of delay was associated with a 12% increase in the odds of clinical sequelae (adjusted odds ratio: 1.12, 95% CI: 1.03–1.22, p = 0.006) (Table 4).
| Adjusted odds ratio | Confidence interval | P value | |||
| Lower | Upper | ||||
| Age | 0.86 | 0.71 | 1.05 | 0.150 | |
| Gender | Male | Ref. | – | – | – |
| Female | 0.44 | 0.07 | 2.75 | 0.379 | |
| Kocher scorea | 0,1, or 2 | Ref. | – | – | – |
| 3 | 1.52 | 0.18 | 12.89 | 0.698 | |
| 4 | 1.11 | 0.13 | 9.41 | 0.922 | |
| Symptoms to surgery (days) | 1.12 | 1.03 | 1.22 | 0.006 | |
3.1 Results of sensitivity analysis
The sensitivity analysis further reinforced the theory that a longer delay to surgery is associated with a higher likelihood of complications. In the sensitivity analysis, where we considered only the admission to surgery delay to mitigate potential reporting bias of duration of symptoms prior to admission, the logistic regression indicated that age, gender, and Kocher scores were not significantly associated with radiologic deformity. However, the admission to surgery delay demonstrated a statistically significant association, with each additional day of surgery delay increasing the odds of radiologic deformity (adjusted odds ratio: 1.96, 95% CI: 1.13–3.41, p = 0.017) (Table S1). Each additional day of delay was also associated with a 21% increase in the odds of clinical sequelae (adjusted odds ratio: 1.21, 95% CI: 1.04–1.42, p = 0.017). The other covariates did not show a significant association (Table S2).
4 Discussion
Septic arthritis of the hip is a challenging diagnosis, particularly in the pediatric population. It is well established that late diagnosis of septic hip arthritis in children is associated with adverse outcomes.3 In this study, we retrospectively evaluated 46 septic hips with late diagnosis and management. In the long-term follow-up, among these patients, 21 (45.65%) hips showed radiological deformity. Moreover, 13 (28.26%) patients had clinical sequelae.
Delayed management is implicated in poor clinical outcomes. In a systematic review, it was revealed that a favorable outcome is achieved in 75% of instances when symptoms manifest for fewer than four days, but this rate drops to 15% when symptoms persist for more than four days.16 Our findings are in line with this statement. Among the 46 cases, only three experienced a delay from symptom onset to surgery of less than four days, and all of them exhibited no deformities during follow-up. In contrast, among those with a longer delay, only 48.8% showed no deformities. A similar pattern emerged when examining the delay from admission to surgery: when the surgery occurred within four days of admission, 77.8% showed no deformity, whereas, for cases with an admission to surgery delay of four days or more, only 15.8% showed no deformity.
Usually, the delays in the management of septic arthritis are due to the challenging nature of making a definite diagnosis, especially in the neonatal and younger pediatric population where localization of infection may be difficult. The main differential diagnosis is transient synovitis of the hip which is classically differentiated by Kocher's criteria.5 Various positive predictive value ranging from 59 to 99.8% has been suggested in circumstances where all criteria are present.5,17 In this study, only 28.26% of the patients had all four classic criteria present. This might be due to the limited number of cases and the fact that most of our study population seeked medical aid late which might have masked the classic presentations. Moreover, we evaluated Kocher's score for possible predictive value in the final clinical and radiological outcomes. Yet, we did not find any meaningful relation between Kocher's criteria and the outcomes. This may be justified by the fact that elevation of Kocher parameters such as white cell count response varies with age in children, generally being elevated in older children, occasionally in younger ones, and rarely in neonates.16
There is uncertainty regarding the precise role of age in the course of septic hips. The age of the patient might play a significant role in the primary clinical diagnosis since at younger ages, the surgeons might face atypical presentations. The culprit microorganism varies in different age groups and this might be another cause of different outcomes. At very young ages, Group-B Streptococcus and gram-negative bacilli, at one to five years old, S. aureus, K. kingae, and S. pneumoniae, and in older ages, S. aureus and N. gonorrhea have been suggested.3,18,19 In the study by Lunseth and Heiple11 younger age, especially under one year of age was associated with poorer radiologic outcomes. In a study of 15 Choi type 4 septic hips, pathology occurred regardless of the patient's age. In the study by Fabry and Meire,14 the results of delayed management in neonates (under 1 year old), especially premature ones, were poor, and it was suggested that these patients are more affected by treatment delay than older ones. Delay in the management of the patient, dislocation, and proximal femoral osteomyelitis were the suggested factors in the Choi type 4 pathogenesis.20 Notably, in our investigation, we observed that increased patient age exhibited an independent albeit modest association with unfavorable radiological outcomes. The age of the patients had a 1.38 odds ratio for radiological outcomes (p value = 0.046). This connection could be attributed to a higher incidence of Staphylococcus aureus as the causative microorganism in older individuals.16S. aureus, is a recognized microorganism associated with poorer outcomes.16 Due to the absence of definitive cultures and the exclusion of microorganism type from our analysis, the impact of microorganism type may have been reflected in the higher age. On the other hand, it should not be forgotten that one of our mortalities which did not enter the analysis, was 1 month old. These findings might also be due to limited case number in our study.
Through the study, we had three cases of mortality. Some studies report no mortality,14,21 however, in a study by Kabak et al. in 14 neonates, 1 case of mortality was reported.22 In our study, one mortality case had significant morbidities and delays in the surgical intervention. This calls for multidisciplinary care in the management of patients with comorbidities and immunocompromised states.
In general, pediatric septic arthritis of the hip can lead to various long-term complications, significantly affecting a child's musculoskeletal health and overall well-being. This study's strength lies in its assessment of different factors and cases with significant delays in presentation or management. We highlighted that delayed surgical management remains the most critical factor influencing the outcomes of patients with septic hips, while the influence of factors like age and laboratory parameters may not be entirely clear-cut or direct. Diligent follow-up, early intervention, and a comprehensive approach addressing orthopedic, growth-related, and psychosocial aspects are essential to minimize the impact of complications and optimize long-term outcomes for affected children.
To the best of our knowledge, this is the most extensive study to date that has examined both the clinical and radiological consequences of delayed management in long-term follow-ups of pediatric septic arthritis of the hip. However, the study is still constrained by the limited number of cases, considering the rare nature of these circumstances.
Author contributions
Seyed Arman Moein, MD (Contributions: Conceptualization and methodology, supervision, project administration, manuscript preparation): Reza Fereidooni, MD (Contributions: data curation, statistical analysis, manuscriptdrafting, and manuscript final editing): Mohammad Hadi Gerami, MD (Contributions: Conceptualization and methodology, study design): Asal Seifaei, MD (Contributions: data collection): Houyar Zarifkar, MD (Contributions: data collection): Amirhossein Kamalinia, MD (Contributions: Conceptualization and methodology, data collection).
All authors reviewed the results and approved the final version of the manuscript.
Ethical considerations
All of the study stages were approved by Ethical Committee of Shiraz University of Medical Sciences (code: IR. SUMS.MED.REC.1402.344). All methods were performed in accordance with relevant guidelines and regulations. Written informed consent was obtained for the publication of two radiographs.
Consent for publication
Not applicable.
Availability of data and materials
The dataset supporting the conclusions of this article is available for academic researchers via the research deputy of Shiraz Medical School (med_thesis@sums.ac.ir) upon reasonable request.
Funding
This study was supported by the Vice-Chancellor for Research, Shiraz University of Medical Sciences, Shiraz, Iran; grant number 28290. The funding source did not influence the study design, patient selection, data collection, analysis, or interpretation, and played no role in the study's authoring.
Guardian/patient's consent
The Institutional Review Board has waived the requirement for written informed consent, provided that no identifying personal patient data are published. Written informed consent was obtained for the publication of two radiographs.
References
- The impact of the current epidemiology of pediatric musculoskeletal infection on evaluation and treatment guidelines. J Pediatr Orthop. 2008;28(7):777-785.
- [Google Scholar]
- Etiology and medical management of acute suppurative bone and joint infections in pediatric patients. J Pediatr Orthop. 1982;2(3):313-323.
- [Google Scholar]
- Septic arthritis of the hip in children: a critical analysis review. JBJS Rev. 2020;8(2)
- [Google Scholar]
- Transient synovitis, septic hip, and Legg-Calvé-Perthes disease: an approach to the correct diagnosis. Pediatr Clin. 2014;61(6):1109-1118.
- [Google Scholar]
- Differentiating between septic arthritis and transient synovitis of the hip in children: an evidence-based clinical prediction algorithm. J Bone Joint Surg Am. 1999;81(12):1662-1670.
- [Google Scholar]
- The use of CRP within a clinical prediction algorithm for the differentiation of septic arthritis and transient synovitis in children. J Bone Joint Surg Br. 2011;93(11):1556-1561.
- [Google Scholar]
- Septic arthritis or transient synovitis of the hip in children: the value of clinical prediction algorithms. J Bone Joint Surg Br. 2010;92(9):1289-1293.
- [Google Scholar]
- Septic arthritis of the neonatal hip: acute management and late reconstruction. J Am Acad Orthop Surg. 2013;21(10):632-641.
- [Google Scholar]
- Update on the management of pediatric acute osteomyelitis and septic arthritis. Int J Mol Sci. 2016;17(6)
- [Google Scholar]
- Surgical treatment of the severe sequelae of infantile septic arthritis of the hip. Clin Orthop Relat Res. 2005;434:102-109.
- [Google Scholar]
- Prognosis in septic arthritis of the hip in children. Clin Orthop Relat Res. 1979;139:81-85.
- [Google Scholar]
- Sequelae of septic arthritis of the hip in children: a new classification and a review of 41 hips. J Pediatr Orthop. 2008;28(5):524-528.
- [Google Scholar]
- Prognostic factors of septic arthritis of hip in infants and neonates: minimum 5-year follow-up. Clin Orthop Surg. 2015;7(1):110-119.
- [Google Scholar]
- Septic arthritis of the hip in children: poor results after late and inadequate treatment. J Pediatr Orthop. 1983;3(4):461-466.
- [Google Scholar]
- Sequelae and reconstruction after septic arthritis of the hip in infants. J Bone Joint Surg Am. 1990;72(8):1150-1165.
- [Google Scholar]
- The management of septic arthritis in children. The Journal of Bone & Joint Surgery British. 2009;91-B(9):1127-1133.
- [Google Scholar]
- Differentiation between septic arthritis and transient synovitis of the hip in children with clinical prediction algorithms. J Bone Joint Surg Am. 2004;86(5):956-962.
- [Google Scholar]
- Current trends of microorganisms and their sensitivity pattern in paediatric septic arthritis: a prospective study from tertiary care level hospital. J Clin Orthop Trauma. 2017;8(1):89-92.
- [Google Scholar]
- What factors contribute to Choi IV sequelae? A retrospective analysis of 15 septic hips. J Clin Orthop Trauma. 2020;11(Suppl 4):S645.
- [Google Scholar]
- Outcomes of paediatric septic arthritis of the hip and knee at 1-20 years in an Australian urban centre. ANZ J Surg. 2019;89(5):562-566.
- [Google Scholar]
- Septic arthritis in patients followed-up in neonatal intensive care unit. Pediatr Int. 2002;44(6):652-657.
- [Google Scholar]