Key results
In this study, we observed a PPC incidence of 8.7% among elderly patients undergoing upper abdominal surgery. One in every five patients with PPCs required transfer to the ICU and reintubation. Based on multivariable analysis, we observed that PCA was associated with a reduced incidence of PPCs in elderly patients following upper abdominal surgery. The result was further validated by a sensitivity analysis using propensity score matching. Additionally, after categorizing patients receiving PCA into different groups, we observed no significant differences in PPCs, unplanned ICU transfers, postoperative LHS, or in-hospital mortality, regardless of the analgesic methods (PCIA vs. PCEA) and opioid drugs (hydromorphone vs. sufentanil).
Interpretation
Risk and beneficial factors for PPCs
The study identified several factors influencing the occurrence of PPCs in elderly patients undergoing upper abdominal surgery. Consistent with previous findings[25], a history of preoperative pulmonary disease was an independent risk factor for PPCs[26]. At the medical center where this study was conducted, elderly patients undergoing upper abdominal surgery are required to undergo routine respiratory mechanics and pulmonary imaging (X-ray or CT) exams. We speculate that in the post-COVID era, the prevalence and detection rate of preoperative pulmonary disease screening will improve, aiding in the accurate preoperative diagnosis of pulmonary diseases, thereby enabling clinicians to take preventive measures against PPCs more promptly. In addition, age has consistently been a risk factor for PPCs[8, 10], with 80 years being a noteworthy threshold where the incidence of PPCs rapidly increases from 7.9–27.3% (≤ 80 vs. >80 years old), emphasizing the imperative to address controllable factors in order to mitigate PPC risk in elderly patients. Some of these factors are indeed controllable and adjustable, such as postoperative analgesia method, surgical duration, surgical approach, and secondary surgery.
Interventions for PPC reduction
Several established interventions may contribute to lowering the incidence of PPCs, including promoting enhanced recovery[27], epidural analgesia[28], prophylactic respiratory physiotherapy[29], goal-directed hemodynamic therapy[30], prophylactic sputum clearance, intraoperative lung-protective ventilation, and postoperative continuous positive airway pressure[31]. The combined use of the first four interventions has been shown to reduce the relative risk of PPCs by 25%[31]. In this study, we focused on the relationship between perioperative pain management strategies and the risk of PPCs.
Given the severity and humanitarian concerns of pain in upper abdominal surgery, 92.6% of elderly patients received PCA. Although there was a significant imbalance in the number of patients between the PCA and non-PCA groups in the entire population, this discrepancy does not imply frailty or more comorbidities in the non-PCA group. In fact, non-PCA patients often declined PCA due to economic reasons (postoperative analgesia not fully covered by medical insurance in China) and fear of side effects of analgesics. To minimize the bias from this imbalance, we matched baseline data for patients in both PCA and non-PCA groups and found that postoperative PCA was still associated with a lower risk of PPCs in elderly patients undergoing major upper abdominal surgery. As previous studies have reported, PCA benefits postoperative recovery in elderly patients by reducing moderate to severe pain, thereby facilitating respiration and early mobilization, and reducing excessive stress hormone levels[32]. Moreover, PCA reduce respiratory muscle dysfunction and enhance ventilatory efficiency[33] to prevent PPCs.
Continuous thoracic epidural analgesia (TEA) was once considered the best analgesic method after abdominal surgery[28, 34], achieving effective analgesia by blocking visceral and somatic pain signals to the spinal cord. However, recent high-quality studies have shown that TEA increases the risk of hypotension and arrhythmias after thoracic surgery[35, 36]. TEA may also impair postoperative lung function by paralyzing respiratory muscles such as intercostal or abdominal muscles[37]. At the center where this dataset was collected, thoracic segment PCEA is mainly used for patients undergoing upper abdominal surgery via laparotomy. However, with the increased adoption of laparoscopic minimally invasive techniques over the years, the proportion of patients experiencing moderate to severe pain after upper abdominal surgery has significantly decreased. Additionally, with the development of multimodal analgesia, auxiliary analgesic measures including local infiltration anesthesia at the wound site, early initiation of oral NSAIDs, traditional Chinese acupuncture, and family-like accompaniment, the demand for traditional opioid analgesics in elderly patients has gradually decreased. Therefore, even low-dose opioid-based patient-controlled intravenous analgesia can achieve effective analgesic effects while reducing interference with respiration and effective coughing[38]. This may explain why there was no difference in the incidence of PPCs between the PCEA and PCIA groups in this dataset whether confounding factors were matched or not, as well as the choice between the water-soluble opioid (hydromorphone) and the lipophilic opioid (sufentanil) in PCIA did not show a significant association with the occurrence of PPCs. From another perspective, compared to the operational difficulty of epidural puncture in elderly patients, PCIA may serve as a more achievable alternative to PCEA for anesthesiologist.
Limitations
The results of this study should be interpreted in light of its limitations. One limitation is the sourcing of patient data from a single center, which limits the extrapolation and scalability of our findings over time and impedes the external validation of the discovered results. Secondly, unobserved variables exist in this study, such as a history of acute respiratory infection within a month before surgery, preoperative pulse oximetry, and postoperative pain intensity. Several strategies were employed to minimize the impacts of these limitations. However, despite these efforts, the inherent limitations of this study could not be completely eliminated. Hence, the existing results should be carefully interpreted in further research.