Reintubation is required for various reasons[3]. Lin P H and colleagues reported that reintubation cases due to accidental removal or self-removal of the tracheal tube both had distinct risk factors and prognoses[8]. In this study, we classified the unplanned reintubation cases by cause, excluded the non-respiratory cases, and mainly focused on AAC-caused reintubation after general anesthesia since the postoperative respiratory compromise cases stood out as the majority (36/48, 75.0%).
Based on our results, age >65 yrs and ASA physical status 3 were identified as highly associated with unplanned reintubation caused by AAC. This result was in accordance with previous results that noted that advanced age and ASA physical status ≥3 should be considered independent risk factors for postoperative respiratory failure[9,10]. Age ≥65 years was also shown to be an important risk factor for failed extubation in ICU patients[11]. For comorbidities, preoperative sepsis was found to have a positive association with reintubation due to AAC, which agrees with the previous literature[2,12]. Heart disease history and cerebral infarction history were shown to be related to AAC-caused unplanned reintubation in the univariable analysis. Heart disease history is in line with previous results that found that patients with underlying chronic cardiac disease are at a high risk for extubation failure[1]. However, some studies reported no increased risk of reintubation in patients with comorbidities such as heart diseases[13], cerebrovascular accidents or central nervous system (CNS) diseases[14,15]. For other comorbidities, a large-scale prospective study suggested that hypertension and insulin-required diabetes mellitus were independent predictors of unanticipated early postoperative intubation[2]. However, in this study, hypertension demonstrated a trend of increased risk in only the univariable analysis, although the result was not statistically significant. Stratifying patients by medication- or insulin-required diabetes mellitus may obtain more accurate results.
A multitude of studies have reported that COPD is highly associated with reintubation[1,2], [14-17]. It was believed that COPD manifested as narrowing of the small airways, leading to an increase in breathing effort and exacerbating respiratory diaphragm muscle fatigue[18]. One possible reason that the results obtained from this study were not in accordance with previous reports is that in the PUMC Hospital, most moderate or severe COPD patients were sent back to the ICU straight after the operations and were thus excluded from the study.
To the best of our knowledge, no previous study has assessed the association between CL grade and reintubation. CL grading is usually used as a predictor for difficult intubation[19,20]. In this study, CL grading was significant in the univariable analysis. We suspect that the crowded pharyngeal structure contributed to the collapsibility of the airway. Regarding laboratory results, a Ccr <70 ml/min was demonstrated to be associated with AAC-caused unplanned reintubation in the univariable analysis. Numerous studies have concluded that chronic kidney disease[14] and renal insufficiency[2, 11, 17] are significant risk factors for reintubation. Although increased WBC counts were identified to be significant in the univariable analysis, they were more likely to be the result of confounding effects, rather than true associations. Hypoalbuminemia was also suggested to be highly associated with postoperative reintubation in some previous studies[11, 17]. However, such a result was not observed in this study.
A number of studies have reported the association between transfusion or RBC transfusion and reintubation[2]. To our knowledge, no previous study has assessed FFP transfusion as a potential risk factor for reintubation. Notably, we found FFP transfusion to be significant in the univariable analysis. Although reintubation patients could not be diagnosed with acute ARDS due to the absence of a blood-gas test, several studies have reported the relationship between FFP transfusion and ARDS[21]. Neto and colleagues found that perioperative FFP transfusion increased the risk of postoperative ARDS[22]. Thus, FFP transfusion may be correlated with ARDS and reintubation after operation. RBC transfusion was not identified as an independent risk factor, consistent with the result from Acheampong D and colleagues[11]. A fluid load ≥ 20 ml kg-1 was revealed as a risk factor in this study; however, other studies found that fluid balance or overload was not a significant risk factor, possibly due to the different definitions of fluid load[15, 23]. There is evidence that an extensive infusion of fluid during an operation results in pulmonary edema and pneumonia[24], which may be correlated with unplanned reintubation.
For the operative-specific factors, head-neck surgery and thoracic surgery were identified as significant risk factors; this result is similar to those in previous reports[15,16]. Of all the unplanned reintubated cases, four thoracic patients were ultrasonically diagnosed with phrenic nerve injury, which is an iatrogenic complication following thoracic and cardiac surgery, with an overall incidence ranging from 1% to 11%[25,26]. In PUMC Hospital, all patients undergoing cardiac operations were extubated routinely in the ICU, therefore, all cardiac cases were excluded from the investigation.
In this study, the incidence rate of reintubation was low compared with that in the previous literature[13,14]. There are mainly four feasible reasons for this variation in the results. First, all the anesthesiologists were trained by the same protocol despite seniority[27]. Second, extubation was routinely conducted by two anesthesiologists, and both the attending and resident doctors were responsible for the case. Third, most critical patients were sent back to the ICU after their operation, which also explained why there were no ASA physical status 4 or 5 patients in the study. Finally, although unplanned reintubation is mandatory to report in the adverse event reporting system, we suspect that there may have been some missing cases.
A small sample size was the major limitation for this study. As identified in the power estimate that was included in the statistical analysis, 60.71% of the potential risk factors had >50% power, and the incidence of reintubation due to AAC was only 0.03%. This severely underpowered analysis may lead to false-negative results. Therefore, the potential factors that were found negative in this study will not necessarily be unrelated to the reintubation caused by AAC. Moreover, sepsis should be considered as an important risk factor for reintubation[2,12], it also revealed statistical significance in the univariable analysis. However, due to the very limited sample size, only two patients were found in the case group while none was found in the control group. The zero event prevented us from including sepsis in the multivariable regression model. A larger sample size may be needed to verify these associations adequately. There were other limitations of this study. For instance, as this was a retrospective case-control study, there might be potential confounders that have causal associations with unplanned reintubation and are unbalanced between the case and control groups, therefore resulting in confounding effects. In addition, data on other potential risk factors, such as respiratory tract infection or hypothermia, which were considered significant in previous studies[15, 17], were not included in this study due to the limitation of the recording system. Prospective, multicenter studies with larger sample sizes and fewer untreated confounders are required to further validate the current conclusions.
One strength of this study was that it specified the cause of reintubation. Therefore, the evaluation of the risk factors had increased accuracy. Additionally, for the first time, we revealed that FFP transfusion and CL grades were significantly related to unplanned reintubation. Finally, all the cases and controls received a combination of intravenous and inhaled anesthesia; to some extent, we prevented the anesthetic method from acting as a confounding factor.
Here, we have attempted to identify risk factors for postoperative unplanned reintubation caused by AAC to prevent unplanned reintubation after general anesthesia. Thus, minimizing patient risk factors, staying alert and making judicious decisions are essential to improve surgical prognoses.