Analysis of risk factors for postoperative wound healing in children with bone cysts Authors: Muhan Yu, Yuxi Su Published: 2026 ABSTRACT AbstractBackgroundBone cysts are common benign bone tumour-like lesions in children and adolescents. Postoperative wound healing complications significantly affect patient outcomes. Current research primarily focuses on cyst recurrence and bone healing, while studies targeting risk factors for poor postoperative wound healing in children are relatively scarce.MethodsThis single-center retrospective cohort analysis included 99 children with bone cysts who were hospitalized and underwent surgical treatment between January 2021 and September 2025. We analyzed these data in November 2025. Univariate analysis and multivariate logistic regression were used to assess the association of demographic, lesion characteristics and surgery-related variables with poor wound healing.ResultsAmong the 99 children, 13 (13.1%) experienced poor wound healing. Multivariate analysis revealed that no or single-dose preoperative antibiotic prophylaxis (OR = 9.808, p = 0.031; OR = 34.161, p = 0.010), physeal involvement of the lesion (OR = 7.764, p = 0.023) and operative time ≥90 min (OR = 8.597, p = 0.026) were independent risk factors for poor surgical wound healing.ConclusionInadequate perioperative antibiotic prophylaxis (either no administration or a single preoperative dose only), physeal involvement of the lesion and prolonged operative time are independent risk factors for poor postoperative wound healing in children. Standardizing antibiotic protocols and identifying high-risk patients based on lesion anatomical characteristics are crucial for reducing the risk of complications. Introduction Bone cysts are common benign bone tumour-like lesions in children and adolescents [1,2], primarily including two types: simple bone cysts (SBCs) and aneurysmal bone cysts (ABCs). Bone cysts are often asymptomatic, and in nearly two-thirds of cases, they are only discovered upon the occurrence of a pathological fracture [3]. Currently, surgical procedures such as curettage with bone grafting and intramedullary decompression remain standard treatments for bone cysts. However, postoperative wound healing complications, such as surgical site infection (SSI), fat liquefaction, wound redness and dehiscence, are common challenges in surgical interventions that critically impact patient outcomes [4]. These complications can lead to treatment failure, necessitate reoperation, prolong recovery, increase healthcare costs and impose significant psychological burden on patients and their families. Within the scope of orthopaedic oncology surgery, factors such as prolonged operative time, extensive resection and the frequent use of bone grafts and internal fixation devices contribute to a relatively higher risk of wound healing problems [5]. Current research on bone cysts, both domestically and internationally, focuses mainly on comparing different surgical techniques for reducing recurrence rates and promoting radiological healing [6,7], often overlooking studies on early postoperative wound complications in children. Most studies involving postoperative wound complications combine bone cysts with other bone tumours for analysis, which may obscure their unique risk profile [5]. Therefore, systematically analyzing the independent risk factors associated with postoperative wound healing in children with bone cysts is of significant clinical importance for preoperative risk assessment and developing targeted preventive strategies. This study aims to identify the independent risk factors affecting postoperative wound healing in children with bone cysts through retrospective analysis. We anticipate that the findings will provide evidence-based guidance for clinicians to preoperatively identify high-risk children, optimize surgical plans and enhance postoperative wound management, ultimately aiming to reduce complication rates and improve overall patient outcomes. Methods Study design and participants This single-center retrospective study was approved by our hospital’s Ethics Committee. We collected the data of children hospitalized and surgically treated for bone cysts at our institution between January 2021 and September 2025. The data were retrospectively analyzed in November 2025. Inclusion criteria were: (1) diagnosis of bone cyst (SBC or ABC) confirmed by both preoperative imaging (X-ray, CT or MRI) and postoperative pathological examination; (2) lesion located in a long bone of the extremities and being primary; (3) first-time surgery; (4) age 0–18 years; (5) postoperative follow-up time greater than 30 days. Exclusion criteria were: (1) concomitant other malignant bone tumours; (2) presence of underlying diseases severely affecting wound healing (e.g. severe malnutrition, diabetes and immunodeficiency disorders); (3) did not undergo surgery; (4) incomplete clinical or follow-up data. Data collection Data were collected from the electronic medical record system, as follows. Demographics: age, sex Clinical characteristics: symptoms before treatment, overweight or obesity status, history of pathological fracture. Lesion characteristics: bone cyst location, pathological type, cyst volume, whether the lesion involved the physis. Surgical data: type of bone graft material, use of internal fixation, operative time (from skin incision to suture), incision length, postoperative drainage tube duration. Perioperative management: perioperative prophylactic intravenous antibiotic usage. Overweight or obesity status was determined according to the international standard for child overweight and obesity established by the International Obesity Task Force (IOTF) in 2000 [8]. Cyst volume was estimated as the maximum length × width × depth on preoperative CT or MRI [9]. As cysts are irregularly shaped, this method yields a volume significantly larger than the actual volume but allows for comparative analysis. The perioperative prophylactic intravenous antibiotic regimen was defined as one of three categories: (1) no antibiotics used; (2) single preoperative dose only; (3) one dose each administered preoperatively and postoperatively. Preoperative/postoperative antibiotics were administered within 24 h before surgery start/after surgery end. Outcome measure The primary outcome was wound condition within 30 days postoperatively. Good healing was defined as primary intention healing. Poor healing was defined as the occurrence of surgical site infection, effusion, haematoma, wound dehiscence, etc. SSI was diagnosed according to the Centers for Disease Control and Prevention (CDC) criteria [10]. Surgical procedure All children underwent surgery under general anaesthesia, involving lesion resection and bone grafting, supplemented postoperatively with brace or cast external fixation. Children with pathological fractures or at high risk of fracture were treated with elastic intramedullary nailing for internal fixation. A longitudinal incision was made over the lesion based on imaging localization, dissecting through skin, subcutaneous tissue, deep fascia and reaching the periosteum via intermuscular planes, with care taken to protect blood vessels and nerves. A cortical window was created at the area of thinned cortex, the lesion was completely curetted and sent for pathological examination and the cyst wall was inactivated with absolute alcohol. The cavity was packed with artificial bone, autogenous bone or allogeneic bone. Satisfactory packing was confirmed under C-arm fluoroscopy. All patients underwent bone grafting; no case relied solely on curettage and blood clot healing. If intramedullary nailing was required, a small incision was made at the distal end of the long bone, the nail was inserted proximally and fixation was confirmed with the C-arm. A negative pressure drain was placed based on the intraoperative situation. The wound was closed, and a cast or brace was applied postoperatively. Perioperative wound management Preoperative skin preparation was performed using 5% povidone‑iodine solution. After wound closure, a combined dressing consisting of nano‑silver dressing and sterile gauze was applied routinely. The dressing was changed every 2–3 days or earlier if soiled, and the incision was cleaned with 1% povidone‑iodine at each dressing change. No routine systemic antibiotics were administered postoperatively beyond the perioperative prophylactic regimen described above. Statistical analysis Data entry and analysis were performed using SPSS software (v.27.0; IBM, Armonk, NY, USA). Continuous variables were compared using independent samples t-test or Mann-Whitney U test. Categorical variables were compared using Chi-square test or Fisher’s exact test. Variables found to be potentially associated with wound healing (p < 0.05) in univariate analysis were included in a multivariate binary logistic regression model to identify independent risk factors. Results are presented as odds ratios (OR) with 95% confidence intervals (CI). A P value < 0.05 was considered statistically significant. Ethical approval This study was approved by the Ethics Committee of the Children’s Hospital of Chongqing Medical University(approval no. 2025-LunShen-LinYan-454). Written informed consent for participation in this retrospective study and for publication of relevant data was obtained from the participants (or, for minors, their legal guardians) upon initial admission. All patient data were anonymized prior to analysis, and this anonymization did not distort the scholarly meaning of the results. This study strictly adhered to the ethical principles established in the Declaration of Helsinki. Results This study ultimately included 99 children with bone cysts who underwent surgical treatment and completed postoperative follow-up. The mean age of the children was 9.6 ± 3.5 years (range: 2–16 years), with a follow-up period of 1.5 to 45 months. Regarding bone graft material, 90 patients (90.9%) received synthetic bone graft, 5 (5.1%) received mixed synthetic and autogenous graft, and 4 (4.0%) received allogeneic graft (Tables 1). Among all included children, 86 (86.9%) had good postoperative wound healing, while 13 (13.1%) experienced poor wound healing. Among these 13 cases with poor healing, three were definitively diagnosed with surgical site infection. Microbiological culture of infectious exudate identified Staphylococcus epidermidis in one case, while two cases were culture-negative. All three of these children were readmitted and received additional therapeutic intravenous antibiotics. The remaining ten cases of poor wound healing presented with a range of non-infectious complications, including periwound fat liquefaction, persistent redness and wound dehiscence (Figure 1). A 9-year-11-month-old boy was admitted with a right humeral bone cyst diagnosed following trauma. Preoperative MRI (A, B) demonstrated a proximal humeral cyst with an old pathological fracture and no physeal involvement. After evaluation, he underwent curettage and synthetic bone grafting without perioperative antibiotics. One month postoperatively, the incision showed purulent discharge, indicating infection. He was readmitted for intravenous antibiotics; pus culture was negative. Univariate analysis revealed that cyst location (p = 0.042), physeal involvement (p = 0.006), perioperative prophylactic antibiotic regimen (p = 0.004), operative time (p = 0.049) and incision length (p = 0.032) were significantly associated with postoperative wound healing status (Tables 1 and 2). The poor healing group had a higher proportion of lesions located in the proximal humerus (53.8%) and a lower proportion in the proximal femur (7.7%). The proportions of cases with physeal involvement and operative time ≥90 min were significantly higher in the poor healing group compared to the good healing group (53.8% versus 16.3% and 84.6% versus 55.8%, respectively). Regarding prophylactic antibiotic administration, the proportion of patients receiving no antibiotics (53.8% versus 44.2%) or a single preoperative dose only (30.8% versus 5.8%) was significantly higher in the poor healing group. The nine patients who received a single preoperative dose of antibiotics were operated by seven different attending surgeons across multiple operative theaters, with no single surgeon or theater dominating these cases. However, factors such as sex, age, overweight/obesity status, history of pathological fracture, type of bone graft material, use of internal fixation, pathological type, drainage duration and cyst volume showed no statistically significant differences between the two groups. Subgroup analysis according to physeal involvement was performed to assess potential confounders (Table 3) Cyst volume was significantly larger in patients with physeal involvement (p = 0.036). No significant differences were observed in incision length (P=0.684) or operative time (p = 0.081). No patient required intraoperative blood transfusion. Table 1.Categorical variables and comparative results. N: number. Χ2: chi-square test, fisher exact. *p < 0.05 was considered significant. **p < 0.01 was considered statistically significantly different. Postoperative wound healing Good healingPoor healing NN%NN%Χ2PSexMale6170.9%861.5%0.4720.717 Female2529.1%538.5% Pathological FractureNo3844.2%430.8%0.8320.362 Yes4855.8%969.2% LocationProximal humerus3034.9%753.8%–0.042* Proximal Femur3641.9%17.7% Others2023.3%538.5% Bone Graft MaterialSynthetic Bone Graft7890.7%1292.3%–0.513 Mixed Bone Graft(Synthetic + Autogenous)55.8%00.0% Allogeneic Bone Graft33.5%17.7% Internal FixationNo3844.2%753.8%0.4250.514 Yes4855.8%646.2% Overweight/ObesityNo5766.3%1076.9%0.5850.655 Yes2933.7%323.1% Involved the PhysisNo7283.7%646.2%9.5360.006** Yes1416.3%753.8% Pathological TypeSBCs7486.0%969.2%2.3570.258 ABCs1214.0%430.8% Prophylactic AntibioticNo3844.2%753.8%10.9450.004** Preoperative55.8%430.8% Preoperatively and Postoperatively4350.0%215.4% Operative Time<90min3844.2%215.4%3.8900.049* ≥90min4855.8%1184.6% Categorical variables and comparative results. N: number. Χ2: chi-square test, fisher exact. *p < 0.05 was considered significant. **p < 0.01 was considered statistically significantly different. Table 2.Continuous variables and comparative results. t: t-test, Z: Mann–Whitney test. SD: standard deviation, M: median, Q1: 1st quartile, Q3: 3rd quartile. *p < 0.05 was considered significant. Postoperative wound healing Good healingPoor healingStatisticPAge(years)9.41 ± 3.4710.92 ± 3.59t = −1.4620.147Incision length(cm)6.00(5.00, 7.28)5.00(4.00, 5.50)Z = −2.1490.032*Drainage tube Duration(days)2.00(2.00, 3.00)2.00(2.00, 3.00)Z = −0.4960.620Cyst Volume(cm3)28.14(13.25, 58.02)18.72(9.14, 100.54)Z = −0.1970.844 Continuous variables and comparative results. t: t-test, Z: Mann–Whitney test. SD: standard deviation, M: median, Q1: 1st quartile, Q3: 3rd quartile. *p < 0.05 was considered significant. Multivariable logistic regression analysis identified that no or single-dose preoperative antibiotic prophylaxis (OR = 9.808, p = 0.031; OR = 34.161, p = 0.010), physeal involvement of the lesion (OR = 7.764, p = 0.023) and operative time ≥90 min (OR = 8.597, p = 0.026) were independent risk factors for poor surgical wound healing. Other variables, including tumour location and incision length, were not independent predictors in the final model in Table 4. Table 3.Comparison of potential confounders between patients with and without physeal involvement. Z: Mann–Whitney test. M median, Q1: 1st quartile, Q3: 3rd quartile. Χ2: chi-square test, fisher exact. *p < 0.05 was considered significant. Involved the Physis NoYesStatisticPIncision length (cm) 6.00(5.00, 7.00)6.00(5.00, 7.52)Z = −0.4070.684Cyst Volume (cm3) 23.21(10.88,56.21)46.20(16.28,101.23)Z = −2.0930.036*Operative Time<90 min35(44.9%)5(23.8%)Χ2 = 3.0480.081 ≥90 min43(55.1%)16(76.2%) Comparison of potential confounders between patients with and without physeal involvement. Z: Mann–Whitney test. M median, Q1: 1st quartile, Q3: 3rd quartile. Χ2: chi-square test, fisher exact. *p < 0.05 was considered significant. Table 4.Multivariate logistic regression analysis. OR superiority ratio, CI confidence interval. *p < 0.05 is statistically different. **p < 0.01 was considered statistically significantly different. BS. EWaldχ2POR (95% CI)Prophylactic AntibioticNo2.2831.0584.6540.031*9.808(1.232-78.069) Preoperative3.5311.3696.6550.010**34.161(2.336-499.616) Preoperatively and PostoperativelyRef.LocationProximal humerus−0.0700.9150.0060.9390.933(0.155-5.602) Proximal Femur−1.7021.3041.7030.1920.182(0.014-2.349) OthersRef.Involved the PhysisYes2.0490.9025.1660.023*7.764(1.326-45.454) NoRef.Operative Time≥90min2.1510.9664.9620.026*8.597(1.295–57.070) <90minRef.Incision length(cm) −0.3240.2152.2870.1300.723(0.475–1.101) Multivariate logistic regression analysis. OR superiority ratio, CI confidence interval. *p < 0.05 is statistically different. **p < 0.01 was considered statistically significantly different. Discussion Surgery is the cornerstone for treating tumour-like lesions in the long bones of children. The management following paediatric bone cyst surgery often faces the risk of postoperative complications, among which surgical site infection is a significant issue leading to repeated surgeries, prolonged hospitalization and increased medical costs. While several studies have analyzed factors influencing wound healing after bone tumour surgery [4,5], few have specifically focused on bone cysts. This study concentrated on identifying independent risk factors for postoperative wound healing in children with bone cysts and found that no or single-dose preoperative antibiotic prophylaxis, physeal involvement and operative time ≥90 min were identified as independent risk factors for poor wound healing. These findings provide key evidence for optimizing perioperative management in such surgeries. Our analysis identified a single preoperative antibiotic dose as an independent risk factor for poor wound healing (OR=34.161, 95% CI: 2.336–499.616)—markedly higher than the OR for receiving no antibiotics. However, this finding is based on only 9 patients (4 events) in the single-dose group, and the wide confidence interval indicates low precision; therefore, it should be interpreted with caution and validated in larger cohorts. Although perioperative antibiotics are commonly used to prevent SSI, the optimal regimen remains debated, and current recommendations for children often reference adult protocols [11]. All children in this cohort received cefazolin [12,13]. Notably, only three patients developed definitive SSI, whereas the primary endpoint “poor wound healing” included both infectious and non‑infectious complications. Consensus on the necessity and optimal duration of prophylactic antibiotics in clean orthopedic surgery is lacking [12,14–17]. A single preoperative dose paradoxically carried a higher risk than no antibiotics. A possible explanation is incomplete suppression, transiently disrupting the local microbiome and immune response without eliminating contaminating bacteria. Our findings support a postoperative supplemental dose to maintain effective tissue concentrations throughout the high-risk period. Multivariate regression analysis revealed that physeal involvement is an independent risk factor for poor postoperative wound healing in children with bone cysts, increasing the risk by 7.8 times (OR = 7.764) compared to lesions not involving the physis. This finding links the risk of early postoperative wound complications to the inherent biological characteristics of the cyst, providing an important anatomical indicator for preoperative risk assessment. The physeal region in skeletally immature children is highly vascularized, physiological characteristic. When a bone cyst abuts or invades the physis, it is essentially an ‘active’ cyst [18]. Haidar et al. [19] strongly demonstrated that ‘active’ cysts within 2 cm of the growth plate, potentially due to persistent communication with it, are resistant to various treatments and very difficult to heal completely. Although their endpoint was radiological cyst healing, the biological behaviour they revealed shares a potential common pathological basis with the wound healing complications we observed. Current research primarily focuses on the relationship between physeal involvement and cyst recurrence [20,21], with scarce literature directly investigating its association with postoperative wound complications. However, the underlying logic is clear: surgery for cysts involving the physis is technically more challenging. Achieving complete curettage necessitates a more extensive surgical field and more thorough stripping of the surrounding periosteum and soft tissues. This inevitably damages the locally rich microvascular network, severely impairing tissue perfusion and repair capacity [19]. Additionally, surgical trauma may induce a stronger local inflammatory response, potentially amplified by the active biological environment of the growth plate itself, collectively leading to increased exudate, fat liquefaction or even wound dehiscence. Our study statistically confirms this association for the first time, emphasizing that physeal status is not only a key predictor of long-term cyst recurrence but also a critical risk predictor for perioperative wound management. Consequently, for children whose preoperative imaging suggests physeal involvement, surgeons should identify them as a high-risk group for wound healing complications. Our subgroup analysis further supports this interpretation. Specifically, although juxtaphyseal lesions had larger cyst volumes (which may imply greater bone graft requirements), incision length and operative time did not differ significantly between patients with and without physeal involvement. Thus, the association between physeal involvement and poor wound healing is not explained by differences in these surgical parameters. Nevertheless, the potential confounding effect of bone graft volume could not be assessed directly, as this variable was not consistently recorded. Furthermore, operative time ≥90 minutes was an independent risk factor for poor wound healing, consistent with the well‑known principle that longer surgery increases tissue trauma and infection risk [5,22,23]. Optimizing surgical efficiency is therefore important to reduce complications in pediatric bone cyst surgery. This study has several limitations. First, its retrospective, single-center design limits generalizability and may introduce unmeasured confounders. Second, the total sample size—particularly only 13 events of poor wound healing—was small, reducing statistical power. Third, we lacked data on surgeon experience, and suture material was available for only 51 of 99 patients (51.5%); all recorded sutures were absorbable, but the high missing rate precluded meaningful analysis. Finally, documentation of bone graft material was not standardized (e.g., “number of boxes” without volume equivalence, varied synthetic grafts and allografts), preventing quantitative adjustment; this remains an unmeasured confounder for physeal involvement. These limitations should be addressed in future prospective studies. Conclusion In summary, an inadequate perioperative prophylactic antibiotic regimen (either no administration or a single preoperative dose only), physeal involvement of the lesion and prolonged operative time are independent risk factors affecting postoperative wound healing in children. The results underscore the importance of standardizing perioperative antibiotic prophylaxis protocols and suggest that the anatomical characteristics of the lesion can serve as a key indicator for preoperative risk assessment. Identifying these risk factors helps clinicians adopt targeted measures during the perioperative period to optimize surgical outcomes and improve patient prognosis.