Comparison of Tools for Postoperative Pulmonary Complication Following Cardiac surgery

doi:10.21203/rs.3.rs-1598014/v1

Download PDF

Research Article

Comparison of Tools for Postoperative Pulmonary Complication Following Cardiac surgery

https://doi.org/10.21203/rs.3.rs-1598014/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Objectives: To review the efficacy of two scoring tools for identifying pulmonary complications after cardiac surgery.

Method: A total of 244 patients underwent elective cardiac surgery between March 2021 and December 2021 were included in this observational study. Three independent physiotherapists use two different sets of PPC scoring criteria, described by Kroenke et al (Kroenke Score) and Reeve et al (Melbourne Group Scale [MGS]) to evaluate PPC status daily following surgery at midday. We then retrospective analysis the occurrence of PPC was performed 14 days postoperatively according to European Perioperative Clinical Outcome (EPCO) definitions (included atelectasis, pleural effusion, pneumonia and respiratory failure).

Result: PPC frequency was 46.31% (113/244) with the Kroenke Score, 25.82% (63/244) with the Melbourne Group Scale. The clinically observed incidence of PPC (atelectasis 41.80%, pleural effusion 24.59%, pneumonia 40.57%, respiratory failure 20.08%) was 53.69% (131/244). The ROC curve shows that the overall validity of the Kroenke score is better than that of the Melbourne Group Scale (AUC, 0.94 vs. 0.88, p=0.01). The Kroenke Score performed better in open-heart surgery cohort than the Melbourne Group Scale (AUC, 0.97 vs. 0.87, p<0.01), while in interventional procedures cohort show similar (AUC, 0.90 vs. 0.86, p=0.32).

Conclusion: The incidence of pulmonary complications following cardiac surgery was highly prevalent. Both the Kroenke Score and the MGS score are effective in identifying patients with postoperative pulmonary complications. The two scoring tools performed comparably in the interventional procedure population. Compared to MGS, the Kroenke Score performed better in the open-heart surgery population.

Clinical Trial Registration: Chinese Clinical Trials Registry, identifier ChiCTR2100054869 (http://www.chictr.org.cn/).

postoperative pulmonary complications

cardiac surgery

hospitalization

cost

physiotherapy

Postoperative Pulmonary Complication (PPC) are well known abnormal perioperative events and all abnormal respiratory conditions in the postoperative phase are classified as such, ranging from self-limiting hypoxemia to life-threatening respiratory failure [1]. The majority of studies use different combinations of several abnormal pulmonary events to describe PPC.

Nowadays, PPC are the most common adverse event after cardiac surgery, which incidence over 50% [2], far exceeding the incidence of cardiovascular complications [3]. For example, pneumonia (2.1–21.6%) [4] and atelectasis (16.6–88%) [5] are the most frequent forms of PPC after cardiac surgery. There are specific risk factors that facilitate the development of PPC in patients undergoing cardiac surgery. Such as the inflammation of the lungs due to general anesthesia and mechanical ventilation; the phenomenon of ischemia-reperfusion in the extracorporeal circulation; sternotomy, and prolonged hypothermia all contribute to impaired pulmonary function [5]. PPC contribute substantially to both high mortality (14–30% mortality at 30 days postoperatively), increased medical costs (over $25,000), prolonged hospital stays (13–17 days longer) and reduced Health-Related Quality of Life for patients, causing a substantial challenge to medical care [6–8].

No standardized criteria for PPC have been established, and the incidence of PPC varies widely among studies [9]. In clinical practice, PPC on the basis of clinical symptoms, physical examination, chest imaging, blood tests, microbiological evidence, etc. Therefore, various diagnostic tools based on clinical evidence have been derived. Kroenke et al. initially created the operational definitions of PPC in order to investigate the impact of severe obstructive pulmonary disease on respiratory outcomes after surgery [10]. This criterion was further modified subsequently to define a clinically significant PPC after cardiac surgery as meeting at least 2 items in Grade 2 or 1 item in Grade 3/4 complications. The Melbourne Group Scale (MGS) was developed to identify PPC most likely to be prophylactic with physical therapy following pulmonary resection via open thoracotomy, such as aspiration and sputum trapped, rather than PPC less susceptible to physical therapy, like pulmonary edema and pleural effusion [11]. Since the first report of PPC, numerous studies have demonstrated the effectiveness of perioperative physiotherapy in reducing the occurrence of PPC [12]. A credible and accessible diagnostic tool would benefit the early screening of PPC and the imposition of physiotherapy, as well as, enhance the comparability of study results.

Therefore, the aims of this study were to compare the current widespread utilization of two scoring tools for the discrimination of PPC after cardiac surgery and to determine the best diagnostic tool.

Design

The Biomedical Ethics Committee of West China Hospital of Sichuan University approved this retrospective observational study of consecutive patients who underwent either cardiac surgery at the Cardiac Surgery Department in West-China Hospital, city of Chengdu, province Sichuan from March 2021 and December 2021. The ethics committee abandoned the program to be reviewed as no intervention modality was involved. The study was strictly undertaken in the principles of the Declaration of Helsinki and reported in compliance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE).

Patient selection

Patients were included for study eligibility only if they received elective cardiac surgery. Demographic data including age, sex, type of surgery and spirometry test were collected. Patients with preoperative respiratory comorbidities (atelectasis, pneumonia, pleural effusion and respiratory failure) and emergency procedure were excluded. Patients were screened by the investigator and interdisciplinary cardiac team (two cardiac surgeons and a physiotherapist) according to echocardiography and chest imaging.

Perioperative medical procedures

Patients will undergo either interventional procedures or open-heart surgery. The choice of procedure was based on the latest guidelines for valvular surgery, the patient's age, relevant comorbidities, and a multidisciplinary team (including a cardiologist, a cardiovascular surgeon, an echocardiography specialist, and an anesthesiologist). All patients were anesthetized with 1.5-2 mg/kg propofol for induction, rocuronium bromide for muscle relaxation, and continuous infusion of propofol and remifentanil for maintenance of anesthesia. All patients were intubated and received positive-pressure ventilation, adjusted by 6–8 ml/kg tidal volume, sustained an inspiration-exhalation ratio of 1:2, adjusted respiratory rate based on body weight and maintained an end-expiratory carbon dioxide concentration of 30–35 mmHg. After surgery, all patients were routinely transferred to the intensive care unit (ICU) with close monitoring. After removal of the airway intubation, all participants will receive a physical therapist consultation for standardized assisted walking training, incentive spirometry and sputum clearance techniques. Antibiotics were initiated when there was clinical and imaging evidence of pneumonic disease, and microbiological culture results were used as a guide for the administration of medication. Branchofiberoscope was started when sputum obstructed the bronchus causing lung lobe collapse. Patients weaning from mechanical ventilation in the ICU were transferred to a general ward and discharged.

Outcomes and clinical definition

A retrospective analysis of the occurrence of PPC was performed 14 days postoperatively. We used the European Perioperative Clinical Outcome (EPCO) definitions recognized PPC (included Atelectasis, Pleural effusion, Pneumonia and Respiratory failure) [13]. However, the definition of respiratory failure (Postoperative PaO2 < 60 mm Hg on room air, a PaO2:FiO2 ratio < 300 mm Hg, or arterial oxyhemoglobin saturation measured with pulse oximetry < 90% and requiring oxygen therapy) as several limitations, such as the high variability of PaO2 at the initiation of inhaled oxygen therapy, PaO2 measurement, which is an invasive procedure and not conducive to consecutive and frequent monitoring. Therefore, we adopted the definition of respiratory failure after surgery by Arozullah et al: use of a ventilator for more than 48 hours or unscheduled reintubation of the trachea [14]. Observed outcomes included need to antibiotic therapy for respiratory infection, time of usage mechanical ventilator, intensive care unit and postoperative length of stay and additional cost in pulmonary complications.

PPC assessment score

Assessment of PPC status was performed daily following surgery at midday by the same three independent physiotherapists using two different sets of PPC scoring criteria, described by Kroenke et al [10] and Reeve et al [15]. The pulmonary complication status during the observation period was evaluated separately by three physiotherapists who blinded to all clinical information and treatment confidential according to different criteria. If the three physiotherapists had the same score for the same patient, that score was used. Otherwise, the median score was selected. If more than one assessment was performed on a specific day, the worst score was used. For the purposes of this paper, these diagnostic scoring tools have been named the ‘Kroenke Score’ and ‘Melbourne Group Scale’, respectively. The variables for each scoring tool are shown in Table 1.

Table 1

Postoperative pulmonary complication (PPC) scoring criteria as described by Kroenke et al and Melbourne Group Scale, respectively.
Kroenke Score	Melbourne Group Scale
Grade 1 Cough, dry. Microatelectasis: abnormal lung findings and temperature 37.5°C without other documented cause; results of chest radiograph either normal or unavailable. Dyspnea, not due to other documented cause. Grade 2 Cough, productive, not due to other documented cause. Bronchospasm: new wheezing or preexistent wheezing resulting in change therapy. Hypoxemia: alveolar-arterial gradient 29 and symptoms of dyspnea or wheezing. Atelectasis: radiological confirmation plus either temperature 37.5°C or abnormal lung findings. Hypercarbia, transient, requiring treatment, such as naloxone or increased manual or mechanical ventilation. Adverse reaction to pulmonary medication. Grade 3 Pleural effusion, resulting in thoracentesis. Pneumonia, suspected: radiological evidence without bacteriological confirmation. Pneumonia, proved: radiological evidence and documentation of pathological organism by Gram stain or culture. Pneumothorax. Reintubation postoperative or intubation, period of ventilator dependence does not exceed 48 hours. Grade 4 Ventilatory failure: postoperative ventilator dependence exceeding 48 hours, or reintubation with subsequent period of ventilator dependence exceeding 48 hours.	1. Chest radiograph report of atelectasis/consolidation. 2. An otherwise unexplained WCC of > 11.2 × 10⁹/L or administration of respiratory antibiotics postoperatively (in addition to those administered routinely postoperatively). 3. Fever as seen by raised oral temperature > 38°C with no focus outside of the lungs. 4. Positive signs of infection on sputum microbiology. 5. Production of purulent (yellow or green) sputum differing from preoperative status. 6. SpO2 < 90% on room air (see measurement protocol below). 7. Diagnosis of pneumonia/chest infection by attending physician. 8. Re-admission to the intensive care unit with problems which are respiratory in origin or a prolonged stay on the intensive care unit (over 36 hours) with problems which are respiratory in origin.
PPC = 2 or more items in the grade 2 complications or 1 item in the grade 3 or 4 complications.	PPC = presence of 4 or more of the 8

Statistical analysis

An independent 2-tailed Student's t-test or Mann-Whitney U-test was used for comparing continuous variables. The continuous data are displayed as mean (standard deviation) and median (25th-75th), while the categorical variables are expressed as numbers (percentages). Each patient needs to be reviewed by 3 physical therapists. Fleiss'κ statistic was used to assess the consistency of results between reporters. Almost perfect agreement was defined if the κ value was greater than 0.80. For better data processing, we defined Grade 1 items as 0, Grade 2 items as 0.5, and Grade 3/4 items as 1 (judged as PPC when the assessment score of 1 is reached) according to the Kroenke Score criteria for diagnosing PPC (satisfying 2 items of Grade2 or 1 item of Grade3/4). Similarly, we define each item as 1 point (judged as PPC when the assessment score is reach at 4 points) based on the MGS criteria for defining PPC (meeting 4 or more items). The sensitivity and specificity of the Kroenke Score and MGS for patient PPC were assessed by applying subject operating characteristic curves (ROC curves). ROC curves for Kroenke Score and Melbourne Group Scale scores were compared using the DELONG method. The proportion of patients reclassified using the Kroenke score and MGS was determined by constructing a reclassification table showing the number of patients assigned to a different category of complexity. We will stratify patients again according to Kroenke Score status (Stage I, score < 1; Stage II, 1 ≤ score < 3; Stage III, score ≥ 3). Statistical tests for all hypotheses were 2-sided and were carried out on the 0.05 level of significance. All statistical analyses as performed by SPSS version 25.0 (SPSS Inc., Chicago, Ill) and MedCalc version 20.0 (MedCalc Software Ltd, Belgium).

Demographics

Two hundred and forty-four consecutive patients were observed; 55.33% (135/244) male, mean age 63.65 years (SD 11.93). Surgical procedures included interventional procedures in (including 127 Transcatheter aortic valve replacement; 3 Transcatheter mitral valve replacement; 9 Transcatheter tricuspid valve replacement; 3 Transcatheter tricuspid valve repair; 1 Thoracic endovascular aortic repair) and open-heart surgery (including 4 Coronary artery bypass grafting [CABG]; 16 Bentall surgery; 38 Mitral valve replacement and Tricuspid valve repair; 2 Aortic valve replacement and Tricuspid valve repair; 14 Double valve replacement and Tricuspid valve repair; 17 Aortic valve replacement; 7 Mitral valve replacement; 2 Tricuspid valve repair; 1 Pulmonary valve repair ) of cases.

Score recognition of different criteria

PPC incidence was 46.31% (113/244) when scored by the Kroenke score, when scored by the Melbourne Group Scale 25.82% (63/244). The incidence of PPC in the study cohort was 53.69% (131/244) as defined by the clinical outcome diagnosis of PPC [Atelectasis 41.80% (102/244), Pleural effusion 24.59% (60/244), Pneumonia 40.57% (99/244), Respiratory failure 20.08% (49/244)].

The Kroenke score was most consistent with the diagnosis of the clinical findings of PPC, especially in the diagnosis of pl0eural effusion and pneumonia, where there was a statistically significant difference in Table 2. The ROC curve also shows that the validity of the Kroenke score is better than that of the MGS (AUC, 0.94 vs. 0.88, p = 0.01) (Fig. 1). A consistent determination of the diagnosis of respiratory failure exists because both scoring criteria include prolonged mechanical ventilation time as one of the factors.

Table 2

Comparison of AUC of the Kroenke Score and Melbourne Group Scale in atelectasis, pleural effusion, pneumonia, and respiratory failure.
Score	Atelectasis	Pleural effusion	Pneumonia	Respiratory failure
Kroenke Score	0.81	0.83	0.92	0.84
Melbourne Group Scale	0.78	0.67	0.82	0.84
P	0.36	< 0.01	< 0.01	0.98

Score recognition of different procedures

The population was categorized according to the procedure. The results showed that there was a large difference in Kroenke Score and MGS for determining PPC in populations undergoing open-heart surgery (AUC, 0.97 vs. 0.87, p < 0.01), while the score performed similarly in interventional procedures cohort (AUC, 0.90 vs. 0.86, p = 0.32).

Hospitalization status

There was a significantly longer ICU and postoperative length of stay (p < 0.01) in patients with PPC as recognized by all scoring tools, and significantly prolonged usage of mechanical ventilation (p < 0.01). The time to use mechanical ventilation, ICU time and length of stay were shorter for patients judged to be PPC by the Kroenke Score compared to the MGS (Table 3). Patients who met the criteria for PPC determination had more additional medical costs and time on antibiotics (Table 3).

Table 3

Hospitalization status for postoperative pulmonary complication (PPC) and None-PPC patients (as determined by Kroenke Score and Melbourne Group Scale).
Group		IV [hours]	NIV [hours]	MV [hours]	ICU [hours]	POS [days]	Cost [thousand CNY]	Antibiotics [days]
Kroenke Score	PPC (113)	12.12 (0.00–20.00)	25.17 (9.92-49.00)	42.00 (18.09–68.97)	60.25 (39.06-86.00)	8.00 (7.00–11.00)	2.52 (1.09–4.14)	0.00 (0.00–5.00)
Kroenke Score	None-PPC (131)	0.00 (0.00-4.53)	0.00 (0.00-16.50)	6.70 (0.00-17.33)	22.67 (19.50-40.42)	6.00 (5.00–8.00)	0.40 (0.00-1.04)	0.00 (0.00–0.00)
Melbourne Group Scale	PPC (63)	14.45 (0.00-45.58)	41.00 (22.17–62.50)	61.25 (37.28–90.83)	70.50 (43.47-120.67)	9.00 (8.00–14.00)	3.68 (2.24–5.45)	4.00 (0.00-8.33)
Melbourne Group Scale	None-PPC (181)	0.00 (0.00-10.42)	0.00 (0.00-17.42)	12.25 (0.00-21.84)	25.33 (20.35–45.21)	6.00 (6.00-7.50)	0.74 (0.00-1.31)	0.00 (0.00–0.00)
IV, invasive ventilation; NIV, noninvasive ventilation; MV, mechanical ventilation; POS, postoperative of stay

Definition of PPC using the Kroenke Score

The maximum value of the Jorden index (the sum of sensitivity and specificity minus 1) in the ROC curve was selected as the cut-off point, and the value of this point was used as the optimal cut-off value for this parameter. It appears that the Kroenke score with two Grade 2 or one Grade 3/4 was the appropriate level for this score to determine PPC. The median postoperative length of stay was 6 days (95% CI 5 to 8) for patients with less than two Grades 2 or one Grade 3/4 compared with 8 days (95% CI 7 to 11) for patients with greater than or equal to two Grades 2 or one Grade 3/4 (P < 0.01). We also found a dose-effect relationship for the assessment of PPC using the Kroenke Score. There were significant differences in the time of mechanical ventilation usage (p < 0.01), the length of stay in the care unit (p < 0.01) and postoperative (p < 0.01), additional medical cost (p < 0.01) and time on antibiotics (p < 0.01) for patients at different stages (Table 4).

Table 4

The time of mechanical ventilation usage, intensive unit care and postoperative length of stay for different stage of postoperative pulmonary complication (as divided by Kroenke Score).
Kroenke Score Stage	IV [hours]	NIV [hours]	MV [hours]	ICU [hours]	POS [days]	Cost [thousand CNY]	Antibiotics [days]
Stage I (131)	0.00 (0.00-4.53)	0.00 (0.00-16.50)	6.70 (0.00-17.33)	22.67 (19.50-40.42)	6.00 (5.00–8.00)	5.38 (4.18–6.41)	0.00 (0.00–0.00)
Stage II (81)	10.17 (0.00-15.88)	16.00 (0.00-34.88)	29.80 (15.16–44.19)	45.13 (24.94–65.46)	7.00 (6.00-8.50)	6.59 (5.96–7.89)	0.00 (0.00-2.33)
Stage III (32)	45.03 (13.06–72.15)	56.92 (39.13–91.65)	89.54 (71.79-172.54)	119.44 (91.95-176.51)	10.50 (8.25–17.50)	15.92 (11.41–23.85)	6.33 (4.67–13.25)
P	< 0.01	< 0.01	< 0.01	< 0.01	< 0.01	< 0.01	< 0.01
IV, invasive ventilation; NIV, noninvasive ventilation; MV, mechanical ventilation; POS, postoperative of stay. Stage I, Kroenke Score < 1; Stage II, 1 ≤ Kroenke Score < 3; Stage III, 3 ≤ Kroenke Score.

Day of PPC manifestation

Most patients presented with PPC as defined by the Kroenke score with two Grade 2 or one Grade 3/4 during the early postoperative period, with only 2.65% patients manifesting late over postoperative days 7 (Fig. 2). The most commonly positive Kroenke score variables in patients with two Grade2 or one Grade3/4 were atelectasis by chest radiological confirmation plus either temperature 37.5°C or abnormal lung findings 94.69%, (107/113), Cough, productive 76.99%, (87/113). Frequency of the other variables was as follows: Cough, dry, 16.81% (19/113); Pleural effusion, resulting in thoracentesis, 14.16%(16/113); suspected pneumonia, 24.78% (28/113); proved pneumonia, 20.35% (23/113); and ventilatory failure, 42.48% (48/113).

The incidence of PPC (atelectasis, pneumonia, pleural effusion, respiratory failure) as determined by clinicians based on commonly used definitions was 131/244 (46.31%). A positive Kroenke Score (score ≥ 1) was associated with increased duration of ventilator usage, length of hospital stay, duration of antibiotic therapy and hospitalization costs.

The MGS was first proposed by Reeve et al. It was originally designed to determine respiratory outcomes following pulmonary resection via open thoracotomy. PPC were identified by eight dichotomous factors and were diagnosed if four or more were met [15]. Under the traditional diagnostic criteria for pulmonary complications, the physician makes the diagnosis based solely on laboratory results and the patient's presenting symptoms, which may be under or over diagnosed. In contrast to traditional clinical diagnostic criteria, MGS takes into account not only preoperative and current pulmonary symptoms, such as sputum, but also clinical decisions regarding the subsequent treatment of adverse pulmonary events, such as whether to stay in the ICU for an extended period of time [15]. Further, Reeve argue that this diagnostic tool is designed to help physiotherapists in their decision making by screening out PPC that may be prevented by physiotherapy, such as pulmonary atelectasis and sputum retention, while PPC that are less suitable for physiotherapy will be eliminated, such as pulmonary edema and pleural effusion11. Therefore, MGS is widely used to determine the occurrence of PPC in thoracic and abdominal surgery [16, 17].

The Kroenke score was created to diagnose pulmonary adverse events after cardiac surgery [10]. Postoperative patients were graded as four levels based on clinical symptoms, laboratory results, chest X-ray findings, and time on ventilator. Compared to MGS, Kroenke adds arterial blood gas analysis results with the clinical decision making after the occurrence of respiratory adverse events. More importantly, the concentration is on changes in clinical symptoms and physical examination, such as conditions that are easily missed by physicians in medical record review, including cough, wheezing, and airway-dilating medications [10]. When only an abnormal radiological presentation was found without significant clinical symptoms or pulmonary auscultation changes (Grade 1), it was considered a subclinical change [18]. Hulzebos et al. first applied the Kroenke Score to review the occurrence of postoperative PPC in patients with CABG. That results showed that this criterion had good discriminatory ability, with statistical differences in length of stay and ventilator use in patients diagnosed with PPC [18]. Chen et al. applied this criterion to patients with cardiac surgery (CABG and/or valvular surgery) patients and was equally effective in identifying patients with PPC [19].

In this study, the two scores were used to determine the ability to identify PPC based on clinical diagnosis by including patients undergoing cardiac surgery. Compared to the Kroenke Score (AUC 0.94, Sensitivity 0.84, Specificity 0.97), the MGS (AUC 0.88, Sensitivity 0.67, Specificity 0.95) was relatively poorly identified. MGS, despite its excellent specificity, is insufficiently sensitive and prone to omit patients with postoperative respiratory adverse events. Therefore, this study classified clinically diagnosed PPC into four common types and judged the recognition ability of two criteria separately. The results showed that the two scoring tools exhibited similar performance when discerning cases of atelectasis versus ventilatory failure. Compared to MGS, the Kroenke Score had better ability to distinguish pleural effusion (AUC 0.83 vs. 0.67, p < 0.01) and pneumonia (AUC 0.92 vs. 0.82, p < 0.01). At the onset of the study, we suspected that the MGS was poor at determining the presence or absence of "pleural effusion". The probable reasons are a) the MGS criteria do not directly describe the clinical symptoms associated with "pleural effusion", the closest description is likely the description of chest imaging [15]; b) The MGS focus on respiratory abnormalities that can be improved by physical therapy, so "pleural effusion" is not included in the list of concerns; c) The incidence of pleural effusion after cardiac surgery has been reported to be 90%, with only 10% of pleural effusions being sufficient to raise symptoms. The most common symptom is dyspnea, while fever and coughing sputum are rarer. Physical examination of the chest reveals diminished breath sounds at the base of the affected lung, and severe pleural effusion can cause hypoxemia and hypercapnia [20]. MGS lacks clinical signs of dyspnea and entries for monitoring partial pressure of oxygen and carbon dioxide; d) Postoperative pleural effusions after cardiac surgery are usually small (less than 25% of one side of the chest), asymptomatic, and nonspecific. This is particularly true for left-sided pleural effusions that appear 1–2 days postoperatively and do not progressively increase [21]. It may not be documented in the medical record due to the absence of serious adverse effects. Therefore, there may be bias when reviewing information through the medical record system.

Interestingly, while the MGS showed excellent performance in its ability to distinguish "pneumonia", the Kroenke Score had incredible power. However, both criteria were used together to determine the occurrence of "pneumonia" through similar entries, including clinical symptoms (cough, fever), imaging (pulmonary solidiation and opacification), non-prophylactic antibiotic use, physician judgment, sputum culture, and laboratory results. Therefore, we speculate that it may be that the two criteria set different thresholds under the same entry. For example, the Kroenke Score sets fever as 37.5℃ and laboratory tests reflecting respiratory conditions as arterial blood gas analysis results alveolar-arterial gradient < 29 or partial pressure of carbon dioxide over 50 mm Hg. However, the MGS defines fever as 38℃ and poorer oxygenation as pulse oximetry (SpO2) < 90%. Fever is common in the days following surgical procedures and 38°C is usually used as a threshold [22]. Hyder et al. analyzed the causes of postoperative fever in 614,525 patients undergoing elective surgery and found that fever due to pneumonia occurred in more than 30% of cases in the early postoperative period (days 1–3) and that infection-related complications became the most common complication in surgical patients from day 4 onwards [23]. Therefore, we assume that there is a good correlation between fever and infection. However, the same way of thought is used in clinical practice, where the use of heat dissipation, such as ice packs or antipyretic drugs, and the prophylactic use of antibiotics postoperatively, usually before the patient develops an excessive temperature, reaching excessive body temperature may be limited [24, 25]. Another possible reason is that the MGS requires ≥ 4 entries to be met when judging pneumonia, whereas the Kroenke Score only requires ≥ 2 items of Grade 2 or ≥ 1 item of Grade 3/4, which is easier and more generalized in the assessment. We suggest that setting too strict diagnostic criteria may lead to underdiagnoses, inadequate antibiotic coverage, and worse prognosis. In general, postoperative hospital-acquired pneumonia should be suspected if the patient presents with clinical signs of infection (e.g., fever, purulent sputum, leukocytosis or leukopenia, decreased oxygen saturation) and radiological imaging showing new infiltrates [26]. On the other hand, arterial blood gas is more accurate as a gold standard for the determination of oxygenation than SpO2, which is susceptible to testing equipment [27]. SpO2 measurement is used in most situations in the clinical circumstance, but may not be reliable. Studies have shown that nearly two-thirds of the alarms triggered by SpO2 monitors are false [28]. In addition, SpO2 is a sluggish indicator that decreases only after a significant decrease in ventilation. Moreover, the use of supplemental oxygen also tends to confound readings, which may lead to delayed interventions and elevated patient safety risks [29]. The MGS assesses the oxygenation by means of SpO2, which may overlook some critically ill patients and result in low testing efficacy.

The results of this study showed that positive events identified by the Kroenke Score mainly occurred 1–3 days postoperatively, which is consistent with the timing of PPC reported by Neto et al [30]. This demonstrates the effectiveness of the Kroenke Score for dynamic assessment and helps to detect PPC occurrence early. However, this study also found certain limitations of the Kroenke Score, such as a high number of positive PPC events and a portion of cases being self-limiting events, such as pleural effusion or mild pulmonary atelectasis [5, 31]. We considered that the current Kroenke Score has good ability to distinguish positive PPC events, but the original diagnostic criteria lacked the ability to stratify PPC severity. Therefore, we re-stratified the population of positive events according to severity. The results showed a significant difference in the respiratory outcomes of interventional versus open-heart surgery, especially in stage1 (69% vs. 32%) and stage3 (6% vs. 45%). This reaffirms that postoperative respiratory adverse events are more frequent and serious in patients undergoing open-heart surgery. Stage 1 patients commonly meet 1 Grade 2 item or multiple Grade 1 items. The Grade 1/2 item set by the Kroenke Score is primarily related to patient symptoms (cough, sputum, dyspnea, and bronchospasm). Patients whose imaging is judged by the clinician as mild atelectasis or pleural effusion may not have remarkable symptoms and are described as "subclinical changes". Therefore, these patients are susceptible to Kroenke Score (-) by the assessors. A higher proportion of interventional patients with stage 1 (less respiratory affected) is more likely to be misdiagnosed compared to open-heart surgery (AUC 0.90 vs. 0.97). To avoid the impact of misdiagnosed results on patients, we have established 3 levels based on the Kroenke Score. The results suggested significant variability in the duration of ventilator usage, length of hospital stay, hospital costs, and duration of antibiotic therapy among the different level of patients. This will help clinicians and physical therapists to determine the extent to which patients' respiratory systems are affected and to facilitate the development of targeted follow-up intervention programs.

It is well known that physiotherapy is effective in improving the postoperative pulmonary status of surgical patients [3]. However, the debate on the necessity and effectiveness of interventions for pleural effusions still exists. Some authors have proposed physiotherapy as part of the management plan for any type of pleural effusion [32, 33], and another group of scholars has argued that physiotherapy interventions are ineffective [11]. Although it is generally acknowledged that non-specific pleural effusion or mild pulmonary atelectasis is a self-limiting disease after surgery, the results of this study show that it also increases patient burden and healthcare costs compared to those who do not develop PPC [34]. Future studies could rely on the Kroenke Score to establish a dynamic grading judgment of the severity of PPC, along with individualized physical therapy programs.

The main limitations of this study include: first, this study is a single-center observational study, and enrolled patients were all from the same medical center, thus limiting the generalizability of the results. Second, all data were collected retrospectively through scoring scales, but three physical therapists performed the assessments separately and analyzed the consistency of results, thus reducing risks of bias. Finally, the data were collected retrospectively and it was not possible to determine assessor satisfaction with the use of the scoring tool, including the time spent on the assessment and the difficulty of the assessment. Further multicenter data inclusion is necessary to establish a prospective study to discern the efficiency of assessment tool use.

The Kroenke Score is a convenient diagnostic tool that makes the determination based on postoperative data and the results can be used by a multidisciplinary team. Compared to MGS, this score is a better approximation of the clinical PPC diagnostic situation and facilitates early intervention.

PPC: Postoperative Pulmonary Complication; MGS: Melbourne Group Scale; ICU: intensive care unit; EPCO: European Perioperative Clinical Outcome; CABG: Coronary artery bypass grafting

Acknowledgments

This study was supported by Sichuan University West China Hospital, Rehabilitation Medicine Center, Key Laboratory of Rehabilitation Medicine in Sichuan Province, Cardiac and Vascular Surgery Department, Anesthesiology Department and Day Surgery Center.

Author Contributions

Y-X Z and Y-Q W conceived, designed, and planned the study. Y-X Z acquisition, analysis, and interpretation of data. Y-XZ and Y-QW interpreted results. Y-X Z, Y-Q G, and W H drafted the report. Y-X Z, Y-Q W, W H, Z-RX L and X Z were involved in the critical revision of the manuscript. W H, Z-RX L and X Z supervision and statistical analysis. Y-Q G, and Y-X Z administrative, technical, or material support. All authors contributed to the article, data acquisition, and approved the submitted version.

Funding

This work was supported financially by the Ministry of Science and Technology of the People's Republic of China (2018YFC20304).

Availability of data and materials

The datasets used and analysed during the current study are available from the corresponding author on reasonable request.

Ethics approval and consent to participate

Consent for publication

All authors take responsibility for all aspects of the reliability and freedom from bias of the data presented and their discussed interpretation. All authors had access to the data and played a role in writing this manuscript. All authors had no financial interests related to this study.

Competing of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Jeong BH, Shin B, Eom JS, et al. Development of a prediction rule for estimating postoperative pulmonary complications. PloS one. 2014;9(12):e113656. https://doi.org/10.1371/journal.pone.0113656
Lagier D, Fischer F, Fornier W, et al. Effect of open-lung vs conventional perioperative ventilation strategies on postoperative pulmonary complications after on-pump cardiac surgery: the PROVECS randomized clinical trial. Intensive care medicine. Oct 2019;45(10):1401–1412. https://doi.org/10.1007/s00134-019-05741-8
Ellenberger C, Schorer R, Bedat B, et al. How can we minimize the risks by optimizing patient's condition shortly before thoracic surgery? Saudi journal of anaesthesia. Jul-Sep 2021;15(3):264–271. https://doi.org/10.4103/sja.sja_1098_20
Wang D, Huang X, Wang H, et al. Risk factors for postoperative pneumonia after cardiac surgery: a prediction model. Journal of thoracic disease. Apr 2021;13(4):2351–2362. https://doi.org/10.21037/jtd-20-3586
Badenes R, Lozano A, Belda FJ. Postoperative pulmonary dysfunction and mechanical ventilation in cardiac surgery. Critical care research and practice. 2015;2015:420513. https://doi.org/10.1155/2015/420513
Brown SR, Mathew R, Keding A, Marshall HC, Brown JM, Jayne DG. The impact of postoperative complications on long-term quality of life after curative colorectal cancer surgery. Annals of surgery. May 2014;259(5):916–23. https://doi.org/10.1097/sla.0000000000000407
Yang CK, Teng A, Lee DY, Rose K. Pulmonary complications after major abdominal surgery: National Surgical Quality Improvement Program analysis. The Journal of surgical research. Oct 2015;198(2):441–9. https://doi.org/10.1016/j.jss.2015.03.028
Fleisher LA, Linde-Zwirble WT. Incidence, outcome, and attributable resource use associated with pulmonary and cardiac complications after major small and large bowel procedures. Perioperative medicine (London, England). 2014;3:7. https://doi.org/10.1186/2047-0525-3-7
Fisher BW, Majumdar SR, McAlister FA. Predicting pulmonary complications after nonthoracic surgery: a systematic review of blinded studies. The American journal of medicine. Feb 15 2002;112(3):219–25. https://doi.org/10.1016/s0002-9343(01)01082-8
Kroenke K, Lawrence VA, Theroux JF, Tuley MR. Operative risk in patients with severe obstructive pulmonary disease. Archives of internal medicine. May 1992;152(5):967–71. https://doi.org/10.1001/archinte.1992.00400170057011
Reeve JC, Nicol K, Stiller K, et al. Does physiotherapy reduce the incidence of postoperative pulmonary complications following pulmonary resection via open thoracotomy? A preliminary randomised single-blind clinical trial. European journal of cardio-thoracic surgery: official journal of the European Association for Cardio-thoracic Surgery. May 2010;37(5):1158–66. https://doi.org/10.1016/j.ejcts.2009.12.011
Batchelor TJP, Rasburn NJ, Abdelnour-Berchtold E, et al. Guidelines for enhanced recovery after lung surgery: recommendations of the Enhanced Recovery After Surgery (ERAS®) Society and the European Society of Thoracic Surgeons (ESTS). European journal of cardio-thoracic surgery: official journal of the European Association for Cardio-thoracic Surgery. Jan 1 2019;55(1):91–115. https://doi.org/10.1093/ejcts/ezy301
Jammer I, Wickboldt N, Sander M, et al. Standards for definitions and use of outcome measures for clinical effectiveness research in perioperative medicine: European Perioperative Clinical Outcome (EPCO) definitions: a statement from the ESA-ESICM joint taskforce on perioperative outcome measures. European journal of anaesthesiology. Feb 2015;32(2):88–105. https://doi.org/10.1097/eja.0000000000000118
Arozullah AM, Daley J, Henderson WG, Khuri SF. Multifactorial risk index for predicting postoperative respiratory failure in men after major noncardiac surgery. The National Veterans Administration Surgical Quality Improvement Program. Annals of surgery. Aug 2000;232(2):242–53. https://doi.org/10.1097/00000658-200008000-00015
Reeve JC, Nicol K, Stiller K, McPherson KM, Denehy L. Does physiotherapy reduce the incidence of postoperative complications in patients following pulmonary resection via thoracotomy? a protocol for a randomised controlled trial. Journal of cardiothoracic surgery. Jul 18 2008;3:48. https://doi.org/10.1186/1749-8090-3-48
Parry S, Denehy L, Berney S, Browning L. Clinical application of the Melbourne risk prediction tool in a high-risk upper abdominal surgical population: an observational cohort study. Physiotherapy. Mar 2014;100(1):47–53. https://doi.org/10.1016/j.physio.2013.05.002
Brocki BC, Westerdahl E, Andreasen J, Andreasen JJ. Can the Melbourne Scoring Scale be used to assess postoperative pulmonary complications in high-risk patients following lung resection? 2018. https://doi.org/10.1183/13993003.congress-2018
Hulzebos EH, Helders PJ, Favié NJ, De Bie RA, Brutel de la Riviere A, Van Meeteren NL. Preoperative intensive inspiratory muscle training to prevent postoperative pulmonary complications in high-risk patients undergoing CABG surgery: a randomized clinical trial. Jama. Oct 18 2006;296(15):1851–7. https://doi.org/10.1001/jama.296.15.1851
Chen X, Hou L, Zhang Y, et al. The effects of five days of intensive preoperative inspiratory muscle training on postoperative complications and outcome in patients having cardiac surgery: a randomized controlled trial. Clinical rehabilitation. May 2019;33(5):913–922. https://doi.org/10.1177/0269215519828212
Light RW. Pleural effusions after coronary artery bypass graft surgery. Current opinion in pulmonary medicine. Jul 2002;8(4):308–11. https://doi.org/10.1097/00063198-200207000-00011
Peng MC, Hou CJ, Li JY, Hu PY, Chen CY. Prevalence of symptomatic large pleural effusions first diagnosed more than 30 days after coronary artery bypass graft surgery. Respirology (Carlton, Vic). Jan 2007;12(1):122–6. https://doi.org/10.1111/j.1440-1843.2006.00972.x
Galicier C, Richet H. A prospective study of postoperative fever in a general surgery department. Infection control: IC. Dec 1985;6(12):487–90. https://doi.org/10.1017/s0195941700063608
Hyder JA, Wakeam E, Arora V, Hevelone ND, Lipsitz SR, Nguyen LL. Investigating the "Rule of W," a mnemonic for teaching on postoperative complications. Journal of surgical education. May-Jun 2015;72(3):430–7. https://doi.org/10.1016/j.jsurg.2014.11.004
Schortgen F, Clabault K, Katsahian S, et al. Fever control using external cooling in septic shock: a randomized controlled trial. American journal of respiratory and critical care medicine. May 15 2012;185(10):1088–95. https://doi.org/10.1164/rccm.201110-1820OC
Bratzler DW, Dellinger EP, Olsen KM, et al. Clinical practice guidelines for antimicrobial prophylaxis in surgery. Surgical infections. Feb 2013;14(1):73–156. https://doi.org/10.1089/sur.2013.9999
Kalil AC, Metersky ML, Klompas M, et al. Management of Adults With Hospital-acquired and Ventilator-associated Pneumonia: 2016 Clinical Practice Guidelines by the Infectious Diseases Society of America and the American Thoracic Society. Clinical infectious diseases: an official publication of the Infectious Diseases Society of America. Sep 1 2016;63(5):e61-e111. https://doi.org/10.1093/cid/ciw353
Gattinoni L, Pesenti A, Matthay M. Understanding blood gas analysis. Intensive care medicine. Jan 2018;44(1):91–93. https://doi.org/10.1007/s00134-017-4824-y
Wiklund L, Hök B, Ståhl K, Jordeby-Jönsson A. Postanesthesia monitoring revisited: frequency of true and false alarms from different monitoring devices. Journal of clinical anesthesia. May-Jun 1994;6(3):182–8. https://doi.org/10.1016/0952-8180(94)90056-6
Ianchulev S, Ladd D, MacNabb CM, Qin L, Marengi N, Freeman J. Use of a Respiratory Volume Monitor to Assess Respiratory Competence in Cardiac Surgery Patients After Extubation. Journal of clinical medicine research. Jan 2017;9(1):17–22. https://doi.org/10.14740/jocmr2033w
Serpa Neto A, Hemmes SN, Barbas CS, et al. Incidence of mortality and morbidity related to postoperative lung injury in patients who have undergone abdominal or thoracic surgery: a systematic review and meta-analysis. The Lancet Respiratory medicine. Dec 2014;2(12):1007–15. https://doi.org/10.1016/s2213-2600(14)70228-0
Touw HR, Parlevliet KL, Beerepoot M, et al. Lung ultrasound compared with chest X-ray in diagnosing postoperative pulmonary complications following cardiothoracic surgery: a prospective observational study. Anaesthesia. Aug 2018;73(8):946–954. https://doi.org/10.1111/anae.14243
Valenza-Demet G, Valenza MC, Cabrera-Martos I, Torres-Sánchez I, Revelles-Moyano F. The effects of a physiotherapy programme on patients with a pleural effusion: a randomized controlled trial. Clinical rehabilitation. Nov 2014;28(11):1087–95. https://doi.org/10.1177/0269215514530579
Moradian ST, Najafloo M, Mahmoudi H, Ghiasi MS. Early mobilization reduces the atelectasis and pleural effusion in patients undergoing coronary artery bypass graft surgery: A randomized clinical trial. Journal of vascular nursing: official publication of the Society for Peripheral Vascular Nursing. Sep 2017;35(3):141–145. https://doi.org/10.1016/j.jvn.2017.02.001
Brookes JDL, Williams M, Mathew M, Yan T, Bannon P. Pleural effusion post coronary artery bypass surgery: associations and complications. Journal of thoracic disease. Feb 2021;13(2):1083–1089. https://doi.org/10.21037/jtd-20-2082

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Comparison of Tools for Postoperative Pulmonary Complication Following Cardiac surgery

Status:

Version 1

Abstract

Figures

Introduction

Methods

Statistical analysis

Result

Discussion

Study Limitations

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Status:

Version 1