Routine blood counts may offer clues to recovery after minimally invasive cardiac surgery, but their predictive value needs careful interpretation. A single-center study of 166 patients found that a higher preoperative platelet-to-lymphocyte ratio (PLR) was associated with postoperative inotropic support lasting more than 24 hours.
The association persisted in several adjusted analyses. However, PLR alone had only modest ability to distinguish patients who would need prolonged support, and the study did not establish a validated risk score or a biomarker-guided treatment strategy.
Study Population and Design
Akkaya and colleagues retrospectively reviewed thoracotomy-based minimally invasive cardiac procedures at Ankara Bilkent City Hospital in Turkey between January 2020 and January 2026. Of 207 patients screened, 166 were included in the descriptive analysis. Procedures included valve operations and atrial septal defect closure. Emergency surgery, concomitant coronary bypass surgery, active infection, preoperative mechanical circulatory support, severe hepatic dysfunction, and missing laboratory data were among the exclusions.
The main adjusted analysis included 154 patients with complete data and 40 endpoint events. It considered age, left ventricular ejection fraction, CPB duration, PLR, and the systemic immune-inflammation index (SII). This was an observational study of associations, not a trial of anti-inflammatory treatment or altered perfusion management.
What Was Measured?
Inflammatory indices from the preoperative blood count
Laboratory measurements were obtained within 24 hours before surgery. PLR was calculated by dividing platelet count by lymphocyte count. SII was calculated as platelet count multiplied by neutrophil count, divided by lymphocyte count. These related indices share platelet and lymphocyte components and therefore convey overlapping information.
Prolonged inotropic support, rather than vasopressor use alone
The primary endpoint was continuous intravenous inotropic therapy continuing beyond 24 hours after surgery, despite adequate volume status and correction of reversible causes of hypotension. Counted agents included dobutamine, dopamine, epinephrine, milrinone, and levosimendan.
Vasopressor-only therapy for vasoplegia or isolated blood-pressure support was excluded. When patients received both types of treatment, classification depended on the duration of the inotropic component. This distinction is important for perfusion and ICU teams: the endpoint was not simply any postoperative vasoactive infusion.
Weaning followed clinical assessment and attending-team judgment, without a biomarker-guided protocol. Prolonged support was therefore an imperfect surrogate for delayed hemodynamic recovery, not a definitive diagnosis of low cardiac output syndrome.
Main Findings
Overall, 41 of 166 patients (24.7%) required prolonged inotropic support. The median PLR was 143 in those requiring prolonged support and 118 in those who did not. Median SII values were 649 and 497, respectively.
| Finding | Reported result | Interpretation |
|---|---|---|
| PLR ≥142 in the combined adjusted model | OR 4.18; 95% CI 1.30–13.40 | Associated with prolonged inotropic support |
| SII ≥462 in the same model | OR 1.66; 95% CI 0.52–5.31 | Not independently significant alongside PLR |
| PLR discrimination | AUC 0.655 | Modest stand-alone performance |
| PLR ≥142 classification performance | Sensitivity 53.7%; specificity 75.2% | Substantial misclassification remained |
| Prolonged support by PLR group | 41.5% with PLR ≥142; 16.8% with PLR <142 | Observed, unadjusted rates in this cohort |
Selected numerical findings from the study, presented in an original IPA comparison. Cutoffs were derived and evaluated in the same dataset and are not externally validated clinical thresholds.
Association does not make PLR a stand-alone predictor
The adjusted odds ratio describes odds, not a fourfold increase in an individual patient’s probability of prolonged support. Its wide confidence interval also indicates considerable uncertainty about the effect size. The modest AUC and sensitivity reinforce that a single PLR result cannot reliably identify every patient who will recover slowly.
Both biomarkers were associated with the endpoint in separate adjusted models. With both entered together as threshold-based variables, PLR remained significant and SII did not. In the combined continuous-variable analysis, both estimates were attenuated. The conclusions therefore depend partly on how these overlapping biomarkers are represented statistically.
Age and bypass duration also mattered
Older age and longer CPB duration were independently associated with prolonged support. The reported adjusted odds ratio for CPB duration was 1.33 per additional 10 minutes. This does not establish that shortening bypass by 10 minutes would reduce risk by a corresponding amount: procedure complexity and other unmeasured factors may contribute to both longer CPB and delayed recovery.
Major Clinical Outcomes
The investigators did not detect statistically significant associations between the inflammatory markers and early mortality, acute kidney injury, or postoperative atrial fibrillation. These findings should not be interpreted as proof that no relationship exists. There were only six deaths and five acute kidney injury events in the entire cohort, limiting the ability to evaluate those outcomes.
Relevance to Perfusion Practice
The study raises a practical question: can readily available preoperative inflammatory indices improve a broader assessment of recovery risk? For perfusion teams, PLR may be a useful research variable to consider alongside ventricular function, planned procedure, operative exposure, and postoperative hemodynamics.
That potential role remains unproven. The study did not test changes to pump flow, pressure targets, cardioplegia, transfusion, inotrope selection, or weaning based on PLR. Neither a high platelet count nor a high PLR establishes a need for antiplatelet or anti-inflammatory treatment.
Because the main adjusted model includes intraoperative CPB duration, it is also not a purely preoperative prediction tool. Prospective work would need to show whether PLR adds useful information beyond established risk factors and whether acting on that information improves outcomes.
Limitations and Evidence Assessment
The retrospective, single-center design limits generalizability and cannot establish causation. The cohort combined procedures with different complexity and expected recovery patterns. Important potential confounders—including renal function, hemoglobin, pulmonary hypertension, established surgical risk scores, valve severity, transfusion, and cardioplegia strategy—were not consistently available for full adjustment.
Only 41 primary events were observed, restricting model complexity and increasing concern about unstable estimates. The thresholds were chosen and tested in the same cohort, which can exaggerate apparent predictive performance. Sensitivity analyses were directionally reassuring, but did not amount to formal internal or external validation.
Finally, clinician-dependent inotropic weaning introduces variability into the endpoint, and inflammatory indices were measured at only one preoperative time point. The most appropriate interpretation is an exploratory association worth testing in independent cohorts, rather than a practice-changing prediction rule.
Source and Attribution
This is an original IPA educational commentary on the findings reported by Akkaya and colleagues. The linked original article is provided unchanged under its Creative Commons Attribution–NonCommercial–NoDerivatives 4.0 license. The cover is an original conceptual illustration, not a microscopic specimen image.
Original Article
Akkaya BB, Gül EB, Işık G, et al. Preoperative platelet-related inflammatory burden is associated with delayed early hemodynamic recovery following minimally invasive cardiac surgery. BMC Cardiovascular Disorders. 2026;26:940. doi:10.1186/s12872-026-06341-3.
