Scholarlink

Background <p>Postoperative liver dysfunction (PLD) is a notable complication following the fixation of intertrochanteric femoral fractures (ITFs) via proximal femoral nail antirotation (PFNA). Beyond increasing patient morbidity and mortality, this condition is also strongly associated with heightened healthcare expenditures. Nevertheless, there remains a lack of a practical, validated predictive model designed specifically for this surgical population. The present study thus seeks to develop and validate a user-friendly nomogram for the early prediction of PLD in ITF patients undergoing PFNA.</p> Methods <p>A single-center retrospective analysis included 390 ITF patients who underwent PFNA between January 2022 and January 2025, selected via predefined inclusion/exclusion criteria. PLD was defined as an albumin–bilirubin (ALBI) score &gt; −2.61 plus a neutrophil-to-lymphocyte ratio (NLR) ≥ 4. Demographic and laboratory data were collected. The Boruta algorithm identified key predictors; a multivariable logistic regression model was built and visualized as a nomogram. Model performance was assessed via receiver operating characteristic (ROC) analysis, calibration curves, and decision curve analysis (DCA), with internal validation using 1000 bootstrap resamples.</p> Results <p>The final model incorporated 7 predictors: preoperative inflammatory markers (NLR, platelet-to-lymphocyte ratio (PLR), monocyte-to-lymphocyte ratio (MLR)), coagulation parameter (activated partial thromboplastin time (APTT)), hematological indicators (hemoglobin, red blood cell count), and Evans fracture classification. Multivariable analysis identified independent PLD risk factors: preoperative APTT (OR = 1.05, <i>p</i> = 0.020), NLR ≥ 4 (OR = 3.90, <i>p</i> &lt; 0.001), PLR &gt; 300 (OR = 10.18, <i>p</i> = 0.026), and Evans Type IV fracture (OR = 3.49, <i>p</i> = 0.019). The nomogram showed strong discriminative power (AUC = 0.771, 95% CI 0.722–0.820), good calibration, a bias-corrected C-index of 0.720, and substantial clinical net benefits via DCA.</p> Conclusion <p>A novel nomogram for predicting PLD in ITF patients after PFNA was successfully developed and internally validated. Using easily accessible clinical variables, this tool enables accurate early risk stratification, helping clinicians identify high-risk patients and optimize peri-operative care. It highlights the importance of inflammatory, coagulation, and fracture-related factors in PLD prediction for this cohort.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Development and validation of a cut-off chart for predicting postoperative liver dysfunction following PFNA fixation: composite endpoint and Boruta-based analysis

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月18日

查看全文
  • Zhu Zhu,
  • Linbing Lou,
  • Luqi Yuan,
  • Le Hu,
  • Jihang Dai

摘要

Background

Postoperative liver dysfunction (PLD) is a notable complication following the fixation of intertrochanteric femoral fractures (ITFs) via proximal femoral nail antirotation (PFNA). Beyond increasing patient morbidity and mortality, this condition is also strongly associated with heightened healthcare expenditures. Nevertheless, there remains a lack of a practical, validated predictive model designed specifically for this surgical population. The present study thus seeks to develop and validate a user-friendly nomogram for the early prediction of PLD in ITF patients undergoing PFNA.

Methods

A single-center retrospective analysis included 390 ITF patients who underwent PFNA between January 2022 and January 2025, selected via predefined inclusion/exclusion criteria. PLD was defined as an albumin–bilirubin (ALBI) score > −2.61 plus a neutrophil-to-lymphocyte ratio (NLR) ≥ 4. Demographic and laboratory data were collected. The Boruta algorithm identified key predictors; a multivariable logistic regression model was built and visualized as a nomogram. Model performance was assessed via receiver operating characteristic (ROC) analysis, calibration curves, and decision curve analysis (DCA), with internal validation using 1000 bootstrap resamples.

Results

The final model incorporated 7 predictors: preoperative inflammatory markers (NLR, platelet-to-lymphocyte ratio (PLR), monocyte-to-lymphocyte ratio (MLR)), coagulation parameter (activated partial thromboplastin time (APTT)), hematological indicators (hemoglobin, red blood cell count), and Evans fracture classification. Multivariable analysis identified independent PLD risk factors: preoperative APTT (OR = 1.05, p = 0.020), NLR ≥ 4 (OR = 3.90, p < 0.001), PLR > 300 (OR = 10.18, p = 0.026), and Evans Type IV fracture (OR = 3.49, p = 0.019). The nomogram showed strong discriminative power (AUC = 0.771, 95% CI 0.722–0.820), good calibration, a bias-corrected C-index of 0.720, and substantial clinical net benefits via DCA.

Conclusion

A novel nomogram for predicting PLD in ITF patients after PFNA was successfully developed and internally validated. Using easily accessible clinical variables, this tool enables accurate early risk stratification, helping clinicians identify high-risk patients and optimize peri-operative care. It highlights the importance of inflammatory, coagulation, and fracture-related factors in PLD prediction for this cohort.