Prediction of acute postoperative pain based on intraoperative nociception level (NOL) index values: the impact of machine learning-based analysis
Authors' conclusion
Our results, even if limited by the small number of patients, suggest that acute postoperative pain is better predicted by a multivariate machine-learning algorithm rather than individual intraoperative nociception variables.
Study context
- Population
- ASA I–III patients aged 18–75 years scheduled for gynecological laparoscopic surgery66 analysed
- Participants
- 70 patients
- Comparison
- Moderate to severe pain vs No moderate to severe pain
- Moderate to severe painn = 42
- Primary endpoint
- Ability of a machine-learning algorithm to predict moderate to severe acute postoperative pain from intraoperative NOL values (cross-validated AUC)
Methodological detailsancillary analysis, machine learning, cross-validated
- Ancillary analysis
- Machine learning
- Cross-validated
Primary result
As reported in the abstract
Prediction of moderate to severe postoperative pain, best algorithm
Penalized logistic regression CV-AUC (best performance)
0.753
0.718–0.788
Further findings
| Outcome | Moderate to severe pain | No moderate to severe pain | P value |
|---|---|---|---|
| NOL post-intubation | 42.8 (interval 31.8–50.6) | 34.8 (interval 25.6–41.3) | P = 0.05 |
| Median NOL during surgery | 13 (interval 11–15) | 11 (interval 8–13) | P = 0.027 |
| Percentage of surgical time with NOL > 25 | 23% (interval 18–18) | 20% (interval 15–24) | P = 0.036 |
| NOL TWA < 10 | 2.54 (interval 2.1–3.0) | 2.86 (interval 2.48–3.62) | P = 0.044 |
| Percentage of surgical time with NOL < 10 | 41% (interval 36–47) | 47% (interval 40–55) | P = 0.022 |
- Incidence of moderate to severe pain
- 63.6% · 42
- NOL post-intubation — ROC AUC
- 0.65 [0.51-0.79]
- Median NOL during surgery — ROC AUC
- 0.66 [0.52-0.81]
- Percentage of surgical time with NOL > 25 — ROC AUC
- 0.66 [0.52-0.79]
- NOL time-weighted average (TWA) < 10 — ROC AUC
- 0.65 [0.51-0.79]
- Percentage of surgical time with NOL < 10 — ROC AUC
- 0.67 [0.53-0.81]
- Incidence of moderate to severe pain: of 66 analysed
Original publication
Prediction of acute postoperative pain based on intraoperative nociception level (NOL) index values: the impact of machine learning-based analysis. J Clin Monit Comput. Published 4 August 2022. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →