A retrospective study of 1,307 patients undergoing elective percutaneous vertebroplasty for osteoporotic vertebral compression fractures found that 136, or 10.4%, had a puncture-site wound event within two weeks of surgery. The composite outcome included erythema or swelling, bleeding or exudate, subcutaneous hematoma, and infection. After adjustment for patient and procedural factors, suture closure was associated with more than twice the risk of the composite outcome, while diabetes, mildly elevated international normalized ratio, and multilevel treatment were also associated with higher risk.
The findings come from consecutive patients treated at Ningde Municipal Hospital in China between January 2022 and June 2026. Patients were excluded if they underwent surgery for a new fracture, had uncontrolled coagulation disorders, received concomitant open spinal surgery, required an intraoperative incision at the puncture site, had malignant tumors or abnormal immune status, or lacked key clinical or follow-up information. The analysis therefore describes elective vertebroplasty performed within one hospital system, rather than all fracture patients or all puncture wounds treated after other forms of spine surgery.
Researchers extracted information from electronic medical records, operative and anesthesia records, laboratory reports, dressing-change notes, wound-clinic assessments, and documented telephone follow-up. All puncture sites were covered with waterproof dressings and assessed within two weeks through inpatient observation, clinical review, or telephone follow-up. A wound event was counted only when explicitly documented in the records. Erythema, swelling, bleeding, exudate, hematoma, and infection were combined into one patient-level outcome, counted once even if a patient experienced several types of event, while individual components were also summarized separately.
Closure was not assigned through a randomized comparison. The operating physician selected nonsuture or suture closure according to intraoperative bleeding, exudation, skin condition, wound tension, and routine practice. Sutured wounds were typically closed with one 3-0 Vicryl Rapide stitch. Among the 1,307 patients, 774 received nonsuture closure and 533 received sutures. An early event was recorded in 53 nonsuture cases, or 6.8%, and in 83 sutured cases, or 15.6%. The crude risk ratio for any event was 2.27, with a 95% confidence interval of 1.64 to 3.15 and a P value below 0.001.
After the researchers adjusted for closure method, diabetes, INR above 1.2, age, sex, body mass index, and multilevel treatment, the association with suturing remained. The adjusted risk ratio was 2.24, with a 95% confidence interval of 1.59 to 3.16 and P below 0.001. Diabetes was associated with an adjusted risk ratio of 1.57, with a 95% confidence interval of 1.11 to 2.21 and P equal to 0.010. An INR above 1.2, defined using the hospital laboratory’s upper reference limit, was associated with an adjusted risk ratio of 1.66, with a 95% confidence interval of 1.08 to 2.57 and P equal to 0.022. Multilevel treatment had an adjusted risk ratio of 1.73, with a 95% confidence interval of 1.03 to 2.89 and P equal to 0.038.
The component findings were concentrated in relatively common superficial manifestations rather than infection. Erythema or swelling occurred in 4.1% of patients with nonsuture closure and 10.9% of those with sutures, producing a crude risk ratio of 2.63, with a 95% confidence interval of 1.73 to 3.99. Bleeding or exudate occurred in 2.7% and 5.8% of the respective groups, with a crude risk ratio of 2.14 and a 95% confidence interval of 1.25 to 3.69. Subcutaneous hematoma was uncommon, affecting two nonsuture cases and three sutured cases, and its wide confidence interval included no difference. Infection was documented in one nonsuture case and no sutured cases, too few events to model infection separately.
Age, sex, and body mass index were not independently associated with the composite outcome in the primary adjusted model. Each additional decade of age had an adjusted risk ratio of 1.15, with a 95% confidence interval of 0.94 to 1.41 and P equal to 0.186. Female sex had an adjusted risk ratio of 0.96, with a 95% confidence interval of 0.66 to 1.39 and P equal to 0.829, while each 1 kg/m² increase in body mass index had an adjusted risk ratio of 1.07, with a 95% confidence interval of 0.99 to 1.17 and P equal to 0.091. These estimates mean that the study did not find clear independent associations for those demographic and anthropometric measures, not that they have no relationship to wound events in every clinical setting.
The suture result is particularly notable because the design cannot establish that suturing caused the additional events. The authors excluded several conditions that could make closure decisions unusual, including uncontrolled coagulation disorders, concomitant open surgery, and wounds requiring an intraoperative incision. Even so, surgeons may have selected sutures for locally difficult wounds, bleeding, exudation, poor skin condition, or excessive tension. Those unrecorded features could confound the comparison, while the fact that sutured patients may have had more wound checks or suture-removal visits could also increase the likelihood that mild redness or exudate was noticed and recorded. The finding should therefore be interpreted as an adjusted observational association rather than proof that sutures themselves produce wound inflammation or other complications.
Several other constraints limit what the study can establish. Follow-up was retrospective and varied among inpatient observation, wound-clinic examination, and telephone reporting, so assessment intensity and recording thresholds were not standardized. Very mild redness or transient bleeding may have been missed, and the records did not consistently capture event duration, patient discomfort, additional care, revisits, or hospitalization. The composite endpoint improved capture of local findings but combined conditions with different causes and severities. Only one infection occurred within two weeks, preventing meaningful analysis of surgical-site infection or later infections. The single-center design and the selected elective population also limit generalizability to open spine surgery, malignant disease, immunosuppression, or puncture sites requiring remedial closure.
Overall, the study identifies four factors associated with the short-term local wound status after elective vertebroplasty: suture closure, diabetes, elevated INR, and multilevel treatment. Its practical signal is that early observation may be especially relevant for patients with these characteristics, although the findings do not establish a clinical rule or a preferred closure technique. Because closure was clinician-selected, follow-up was not standardized, and the endpoint combined mild events with rare infection, the results point to priorities for further prospective research rather than definitive guidance. Within those limits, the observed 15.6% event rate after suturing compared with 6.8% after nonsuture closure provides the clearest signal in the cohort, while the observational design makes the mechanism unresolved.
Subject of Research: Factors associated with early puncture-site wound events after percutaneous vertebroplasty for osteoporotic vertebral compression fractures
Article Title: Factors associated with early puncture-site wound events after percutaneous vertebroplasty for osteoporotic vertebral compression fractures: a retrospective cohort study
Article References: He, X., Huang, C., Zheng, B., Wu, X., Huang, C., Lu, S., Huang, Y., Chen, J., Huang, J., & Li, B. (2026). Factors associated with early puncture-site wound events after percutaneous vertebroplasty for osteoporotic vertebral compression fractures: a retrospective cohort study. Frontiers in Medicine, 13, Article 1988843. https://doi.org/10.3389/fmed.2026.1988843
Image Credits: AI Generated
DOI: 10.3389/fmed.2026.1988843
Keywords: Percutaneous vertebroplasty, Osteoporotic fractures, Puncture-site wounds, Suture closure, Diabetes, Coagulation, Multilevel treatment, Retrospective cohort, Postoperative care, Factors, associated, early
Login
