Wang et al.

Journal of Cardiothoracic Surgery (2024) 19:309 Journal of

https://doi.org/10.1186/s13019-024-02797-6
Cardiothoracic Surgery

RESEARCH Open Access

A nomogram predicting pneumonia

after cardiac surgery: a retrospective modeling
study
Kuo Wang1†, Hai‑Tao Zhang2†, Fu‑Dong Fan3, Jun Pan3, Tuo Pan2 and Dong‑Jin Wang1*

Abstract
Background Postoperative pneumonia (POP) is the most prevalent of all nosocomial infections in patients who
underwent cardiac surgery. The aim of this study was to identify independent risk factors for pneumonia after cardiac
surgery, from which we constructed a nomogram for prediction.
Methods The clinical data of patients admitted to the Department of Cardiothoracic Surgery of Nanjing Drum Tower
Hospital from October 2020 to September 2021 who underwent cardiac surgery were retrospectively analyzed,
and the patients were divided into two groups according to whether they had POP: POP group (n=105) and non-
POP group (n=1083). Preoperative, intraoperative, and postoperative indicators were collected and analyzed. Logistic
regression was used to identify independent risk factors for POP in patients who underwent cardiac surgery. We
constructed a nomogram based on these independent risk factors. Model discrimination was assessed via area
under the receiver operating characteristic curve (AUC), and calibration was assessed via calibration plot.
Results A total of 105 events occurred in the 1188 cases. Age (>55 years) (OR: 1.83, P=0.0225), preoperative malnu‑
trition (OR: 3.71, P<0.0001), diabetes mellitus(OR: 2.33, P=0.0036), CPB time (Cardiopulmonary Bypass Time) > 135
min (OR: 2.80, P<0.0001), moderate to severe ARDS (Acute Respiratory Distress Syndrome )(OR: 1.79, P=0.0148), use
of ECMO or IABP or CRRT (ECMO: Extra Corporeal Membrane Oxygenation; IABP: Intra-Aortic Balloon Pump; CRRT:
Continuous Renal Replacement Therapy )(OR: 2.60, P=0.0057) and MV( Mechanical Ventilation )> 20 hours (OR: 3.11,
P<0.0001) were independent risk factors for POP. Based on those independent risk factors, we constructed a sim‑
ple nomogram with an AUC of 0.82. Calibration plots showed good agreement between predicted probabilities
and actual probabilities.
Conclusion We constructed a facile nomogram for predicting pneumonia after cardiac surgery with good discrimi‑
nation and calibration. The model has excellent clinical applicability and can be used to identify and adjust modifiable
risk factors to reduce the incidence of POP as well as patient mortality.
Keywords Cardiac surgery, Postoperative pneumonia, Nomogram, Outcomes

†
Kuo Wang and Hai-Tao Zhang contributed equally to this work.
*Correspondence:
Dong‑Jin Wang
dongjin_wang@126.com
Full list of author information is available at the end of the article

© The Author(s) 2024. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which
permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or
other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory
regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this
licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecom‑
mons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
Wang et al. Journal of Cardiothoracic Surgery (2024) 19:309 Page 2 of 10

Introduction Drum Tower Hospital. All patients included in this study

Postoperative pneumonia (POP) is the most prevalent provided written informed consent. Inclusion criteria
of all nosocomial infections in patients who underwent were patients the ages of 18 and 80 who underwent open-
cardiac surgery [1–3]. Postoperative pneumonia (POP) heart surgery, including aortic valve replacement (AVR),
is the most common infection after cardiac surgery, mitral valve replacement (MVR), MVR+AVR, aortic
with a prevalence ranging from 2% to 10%, especially surgery + AVR, isolated coronary artery bypass grafting
in the first postoperative week [1–7]. Age, smoking his- (CABG), valve +CABG surgery, thoracic aortic surgery
tory, duration of mechanical ventilation (MV) are well- and others. Exclusion criteria included definite preop-
known risk factors that affect POP. Studies have shown erative infection, such as a preoperative temperature of
that multiple drug-resistant pathogens can lead to POP, ≥38°C or a white blood cell count above the upper limit
which makes it more difficult to treat, prolongs hospi- of normal, missing perioperative data, and infection due
talization, and increases the mortality rate of patients to causes other than pneumonia (eg, isolated surgical site
who develop pneumonia by 5 to 17 times compared with infection and isolated urinary tract infection).
non-pneumonia patients [4–7]. Therefore, many studies Those who are in violation of medical ethics, have con-
have focused on identifying risk factors for pneumonia founding factors that may seriously affect the results, or
after cardiac surgery [8–10]. Understanding the risk fac- have poor compliance should be excluded. Those with
tors for POP allows for the development of more effec- other research diseases that may have a significant impact
tive prevention strategies, which are important guides for on the results of the study or the life and health of the
early identification, targeted prevention and treatment, patient. The risk of benefit to patients, such as adverse
thereby improving postoperative recovery and quality of events, should be weighed. Whether a specific age group
life for cardiac surgery patients. can cause bias in trial results. The age range selected for
Many risk factors have been reported repeatedly, this study was 18-80 years. Natural persons under the age
including advanced age, hypoalbuminemia, hypertension, of eighteen are minors, whose cardiorespiratory develop-
smoking history, diabetes mellitus, poor cardiac function ment is immature and biases the experimental results.
(NYHA class III-IV), BMI ≥ 24 kg/m2, previous cardiac In addition, minors can on whether to participate in the
surgery, cardiopulmonary bypass time (CPB time) > 120 experiment, not only to obtain their own consent, but
minutes, blood transfusion [8–10]. Based on these risk also to obtain their guardian’s informed consent and sign
factors, previous studies have constructed some predic- the informed consent form, resulting in less enrollment
tive models to assess the risk of pneumonia in patients data. Older patients may be in poorer general condi-
after cardiac surgery. However, the emergence and prev- tion, have more concomitant diseases or co-morbidities,
alence of resistant bacteria significantly increases the resulting in increased study-related risks and a greater
risk of POP, and patient characteristics also vary [11]. likelihood of adverse events. In addition, the high prev-
In addition, with the tremendous advances in surgical alence of cognitive impairment in the elderly makes
and anesthetic techniques, the baseline and comorbidity it difficult to recruit, to be fully informed, and to make
characteristics of patients have changed greatly in recent independent decisions about whether to participate in a
years. In addition, malnutrition is prevalent among hos- clinical study, and therefore fewer data are available.
pitalized patients with heart failure, significantly increas-
ing the risk of readmission in these patients [12], but few Data collection
POP-related studies have incorporated this into predic- To investigate the risk factors for pneumonia after car-
tive models. Therefore, there is still a need for an up-to- diac surgery, we examined preoperative, intraoperative,
date study of risk factors for pneumonia after cardiac and postoperative indicators. Preoperative indicators
surgery. including age, gender, body mass index, malnutrition,
The aim of this study was to identify independent risk smoking history, hypertension, diabetes mellitus, stroke,
factors for pneumonia after cardiac surgery, from which myocardial infarction, estimated glomerular filtration
we attempted to construct a nomogram for prediction. rate, sequential organ failure assessment score, left ven-
tricular ejection fraction (%),NYHA class. Intraopera-
Methods tive indicators including emergency surgery, procedure
Study design and participants name, minimally invasive, duration of CPB (min), deep
The data of this study come from the database of patients hypothermia circulatory arrest, transfusions (%). Post-
admitted to the Department of Cardiothoracic Surgery of operative indicators including moderate to severe ARDS,
Nanjing Drum Tower Hospital who underwent cardiac use of ECMO or IABP or CRRT, acute kidney injury,
surgery from October 2020 to September 2021 [13]. This mechanical ventilation time, length of intensive care unit
study was approved by the Ethical Committee of Nanjing stay (days), death.
Wang et al. Journal of Cardiothoracic Surgery (2024) 19:309 Page 3 of 10

Definition of important variables by intractable hypoxemia. Oxygen levels in the blood

We adopted the GLIM criteria as the standard for assess- can be divided into three categories by comparing the
ing malnutrition in adult hospitalized patients, and the level of oxygen in the blood and the amount of oxy-
main element of the GLIM criteria is that the assess- gen needed to be given to achieve that level: Mild 200
ment of malnutrition is clearly divided into two steps: mmHg < PaO2/FiO2 ≤ 300 mmHg, PEEP or CPAP ≥
"nutritional screening" and "diagnostic assessment". The 5 cmH2O, and possibly noninvasive ventilation in the
first step is nutritional screening, which emphasizes the mild ARDS group; Moderate: 100 mmHg < PaO2/
use of clinically validated nutritional screening tools. FiO2 ≤ 200 mmHg, PEEP ≥5cmH2O; Severe: PaO2/
In the second step, based on a positive screening, the FiO2≤100mmHg, PEEP≥5cmH2O. Note: FiO2: inspired
patient is then assessed for malnutrition and graded for oxygen concentration; PaO2: partial pressure of arterial
its severity. Smoking history referred to previous daily oxygen; PEEP: positive end-expiratory pressure; CPAP:
or current smoking. Hypertension referred to previous continuous positive airway pressure.
diagnosis, using antihypertensive medication, or blood Extracorporeal membrane oxygenation (ECMO) is
pressure ≥140/90 mmHg. Diabetes is diagnosed when a to draw venous blood out of the body and pump the
patient has a fasting blood glucose level of 7.0 mmol/L human oxygenator outside the body, and the oxygen-
or more, or a blood glucose of 11.1 mmol/L or more two ated blood is discharged from carbon dioxide and then
hours after a meal. Stroke is a general term for acute cer- infused back into the body, replacing the cardiopul-
ebrovascular disease, which is a type of cerebral blood monary function and gaining valuable time for rescue
circulation disorders with sudden fainting and uncon- treatment. Intra-Aortic Balloon Pump (IABP) is one of
sciousness, accompanied by crooked mouth, unfavorable the left ventricular assist devices, which can improve
speech and hemiplegia as the main symptoms. Myocar- the balance of oxygen supply and demand of myo-
dial infarction (MI) is defined as ischemic necrosis of the cardium, increase cardiac output, increase ischemic
myocardium, which is based on coronary artery disease, myocardial perfusion, and promote left ventricular
in which blood flow to the coronary arteries is drastically function by increasing coronary blood flow and reduc-
reduced or interrupted, resulting in severe and prolonged ing cardiac afterload. Continuous renal replacement
acute ischemia of the corresponding myocardium, which therapy (CRRT) is a treatment that replaces the kidneys
ultimately leads to ischemic necrosis of the myocardium. to remove metabolites and toxins and correct water,
The SOFA score, also known as the Sequential Organ electrolytes, and acid-base imbalances. We put these
Failure Score, is a scoring system for making healing together considering that these are post-operative inva-
judgments about a patient, which involves determining sive procedures that they have a lot in common. The
the degree of impairment of major organ function. The use of these operations can significantly improve the
scoring system is divided into six main sections, namely cardiac function status of patients after cardiac surgery,
respiratory function, coagulation function, liver func- maintain hemodynamic stability, improve systemic tis-
tion, cardiac system, central nervous system and renal sue perfusion, reduce the amount of vasoactive drugs,
function, with scores ranging from 0-4. The level of the and improve the probability of smooth withdrawal and
score predicts the mortality rate of septic patients, with discharge survival of patients. Accurate timing of these
the mortality rate increasing by more than 50% for every procedures, enhanced clinical monitoring, and reduced
30% increase in the score. Other types of cardiac surgery complications are key to treatment. Acute kidney injury
include congenital heart disease (such as ventricular sep- (AKI) was defined as a rise in serum creatinine of ≥26.5
tal defect and atrial septal defect), atrial myxoma, and tri- μmol/L within 48 hours or a rise in creatinine to ≥1.5
cuspid valve replacement, but these procedures are less times the baseline value or a urine output of <0.5 ml/
common in this study. Minimally invasive heart surgery kg-h for 6 hours within 7 days.
involves a small incision (i.e. an incision) through the
chest. In this way, the surgeon is able to reach the heart
through the ribs. Blood transfusion is the importation of Study outcomes
red blood cells, platelets and cold precipitates. The unit The primary outcome of the study was pneumonia after
of blood transfusion is % is the volume of blood trans- cardiac surgery. Pneumonia was diagnosed when the
fused divided by the total body weight. patient meets both clinical and bacteriological strategies.
Acute Respiratory Distress Syndrome (ARDS) is an (1): Clinical signs: The presence of a new or progressive
acute, diffuse, inflammatory lung injury resulting in radiographic infiltrate plus at least two of three clinical
increased pulmonary vascular and epithelial perme- features (fever greater than 38°C, leukocytosis or leuko-
ability, pulmonary edema and gravity-dependent pul- penia, and purulent secretions) [14, 15]. (2): Pathogenic
monary atelectasis, and a clinical syndrome marked bacteria were detected twice in sputum culture.
Wang et al. Journal of Cardiothoracic Surgery (2024) 19:309 Page 4 of 10

Statistical analysis the corresponding scores, the individualized probability

Analyses were performed using R version 4.2.1. Cat- of POP after cardiac surgery can be directly and easily
egorical variables were summarized as frequencies (%) predicted. The nomogram demonstrated excellent dis-
and compared using the χ2 or the Fisher’s exact tests, criminate power, with an AUC of 0.82(95% CI 0.78–0.87)
as appropriate. Continuous variables were expressed as (Fig. 2), which confirms the good utility of the model in
median (interquartile range (IQR)) and compared using predicting the development of POP after cardiac sur-
Mann-Whitney U test. A multivariable logistic regression gery. To examine the goodness-of-fit of the model, the
model was used to assess the independent value of pre- calibration plot was used. Calibration plots showed good
dictor variables. For this analysis, quantitative variables agreement between predicted probabilities and actual
were categorized establishing optimum cutoff thresholds probabilities (Fig. 3).
selected from their ROC curves. A backward stepwise Of interest to us, we found that malnutrition was one
approach was followed, including as candidate variables of the most important risk factors for POP. We divided
all those that showed univariate significance better than the patients into two groups according to whether they
P<0.05. Discrimination was measured using the area were malnourished before surgery. We found that the
under the receiver operating characteristic curve (ROC) incidence of pneumonia was significantly higher in mal-
and its 95% CIs. All reported P values are two sides, and nutrition patients than in patients without malnutrition
values of P<0.05 were considered to indicate statistical (Fig. 4).
significance.
Discussion
Results POP is now considered to be a significant cause of mor-
During the study period, a total of 1188 patients who bidity and mortality in cardiac surgery patients, and this
underwent cardiac surgery were included. The incident study further confirms this. In our patient cohort, more
of POP was 8.8% (105/1188 patients). And mortality rate than half of the infections occurred within 48 hours after
of patients with POP was 15.2% (16/105 patients), signifi- the procedure. Infected patients had a significantly higher
cantly higher than that of patients without POP (OR:17.5, mortality rate (1.9% vs. 14% P<0.001) and prolonged hos-
95%CI (8.0-39.9); P<0.001). pital stay. Significant increases in mortality and other
The univariate analysis showed that preoperative poor prognosis in patients with POP, consistent with
related factors included age, malnutrition, diabetes the results in the literature, underscore the importance
mellitus , stroke ,myocardial Infarction (MI), eGFR < of identifying predictors and high-risk patients [7–9].
60ml/min, SOFA score ≥ 1 and NYHA class III or IV ; Therefore, it is important to diagnose POP as early as
intraoperative correlates included emergency surgery, possible.
AVR or MVR, Valve +CABG surgery, thoracic aor- In this cohort study, we evaluated the value of preop-
tic surgery, duration of CPB (min), deep hypothermia erative, intraoperative, and postoperative indicators for
circulatory arrest(DHCA) and transfusions (%); and predicting POP. We identified that age, malnutrition,
postoperative correlates included moderate to severe diabetes mellitus, CPB, moderate to severe ARDS, use of
ARDS ,Use of ECMO or IABP or CRRT ,acute kidney ECMO or IABP or CRRT and MV time were independ-
injury(AKI ),mechanical ventilation (MV) > 48 hours, ent risk factors for POP. Finally, we constructed a sim-
duration of MV (hours) and length of ICU stay (days), ple nomogram and achieved good discrimination and
all of which were statistically significant risk factors for calibration.
postoperative pneumonia. (Table 1). This work was based on numerous efforts to optimize
Multivariate analysis of risk factors for pneumonia after perioperative management in cardiac surgery.
cardiac surgery was showed in Table 2, age (>55 years) A single-center study including 6,222 patients con-
(OR: 1.83, P=0.0225), preoperative malnutrition (OR: cluded that advanced age, chronic lung disease, periph-
3.71, P<0.0001), diabetes mellitus (OR: 2.33, P=0.0036), eral arterial disease, and CPB time >100 minutes,
CPB > 135 min (OR: 2.80, P<0.0001), moderate to severe intraoperative red blood cell infusion, and preoperative
ARDS (OR: 1.79, P=0.0148), use of ECMO or IABP or and intraoperative application of IABP were risk factors
CRRT (OR: 2.60, P=0.0057) and MV > 20 hours (OR: for postoperative pneumonia, and a scoring system was
3.11, P<0.0001) were independent risk factors for POP. developed accordingly. The AUC of the final training set
Based on a multivariate logistic regression model, a was 0.72, but this retrospective study did not include bio-
nomogram model was developed to predict the probabil- markers [10]. Another study using 13,380 patients from
ity of POP after cardiac surgery (Fig. 1). The coefficients four centers constructed a model with 10 preoperative
of these variables were adjusted to a range of scores from and intraoperative risk factors including history of smok-
0 to 100, reflecting their relative importance. By summing ing, diabetes, chronic obstructive pulmonary disease,
Wang et al. Journal of Cardiothoracic Surgery (2024) 19:309 Page 5 of 10

Table 1 Univariate analysis of risk factors of pneumonia after cardiac surgery. Data are n (%) or median (IQR)
No pneumonia Pneumonia P value
(n=1083) (n=105)

Preoperative risk factors

Age (year) 58.00 (50.00, 67.00) 63.00 (56.00, 71.00) <0.001
Gender (male) 653 (60.3) 68 (64.8) 0.43
BMI (kg/m2) 23.95 (21.94, 26.36) 24.03 (21.37, 27.10) 0.64
Malnutrition 72 ( 6.6) 28 (26.7) <0.001
Smoking history 137 (12.7) 13 (12.4) 1
Hypertension 513 (47.4) 60 (57.1) 0.07
Diabetes mellitus 134 (12.4) 25 (23.8) 0.002
Stroke 78 ( 7.2) 16 (15.2) 0.006
MI 126 (11.6) 30 (28.6) <0.001
eGFR < 60ml/min 67 ( 6.2) 16 (15.2) 0.001
SOFA score ≥ 1 457 (42.2) 57 (54.3) 0.022
LVEF (%) 55.00 (51.00, 56.00) 55.00 (49.00, 57.00) 0.297
NYHA class III or IV 665 (61.4) 88 (83.8) <0.001
History of cardiac surgery 68 ( 6.3) 8 ( 7.6) 0.744
Operative risk factors
Emergency surgery 115 (10.6) 27 (25.7) <0.001
Procedure name
Isolated CABG 159 (14.7) 9 (8.6) 0.117
AVR or MVR 416 (38.4) 29 (27.6) 0.038
AVR + MVR 131 (12.1) 6 (5.7) 0.073
Valve +CABG surgery 56 (5.2) 17 (16.2) <0.001
Thoracic aortic surgery 239 (22.1) 36 (34.3) 0.007
Others 82 (7.6) 8 (7.6) 1
Minimally invasive 194 (17.9) 10 (9.5) 0.041
Duration of CPB (min) 123.00 (85.50, 166.00) 165.00 (126.00, 209.00) <0.001
DHCA 204 (18.8) 35 (33.3) 0.001
Transfusions (%) 0.50 (0.00, 1.70) 1.50 (0.50, 2.80) <0.001
Postoperative risk factors
Moderate to severe ARDS 337 (31.1) 62 (59.0) <0.001
Use of ECMO or IABP or CRRT​ 32 ( 3.0) 23 (21.9) <0.001
AKI 80 (7.4) 22 (21.0) <0.001
MV > 48 hours 52 (4.8) 36 (34.3) <0.001
Duration of MV (hours) 8.00 (5.00, 16.00) 19.00 (9.50, 95.00) <0.001
Length of ICU stay (days) 2.00 (2.00, 3.00) 5.00 (3.00, 13.00) <0.001
Death 11 (1.0) 16 (15.2) <0.001
BMI Body Mass Index, MI Myocardial Infarction, eGFR estimated Glomerular Filtration Rate, SOFA Sequential Organ Failure Assessment, LVEF Left Ventricular Ejection
Fraction, NYHA The New York Heart Association Functional Classification, CABG Coronary Artery Bypass Graft, AVR Aortic Valve Repair or Replacement, MVR Mitral Valve
Repair or Replacement, CPB CardioPulmonary Bypass, DHCA Deep Hypothermia Circulatory Arrest, ARDS Acute Respiratory Distress Syndrome, ECMO ExtraCorporeal
Membrane Oxygenation, IABP Intra-Aortic Balloon Pump, CRRT​ Continuous Renal Replacement Therapy, AKI Acute kidney injury, MV Mechanical Ventilation, ICU
Intensive Care Unit

renal insufficiency, age (years), left ventricular ejection multiple organs in the body by prolonging the duration of
fraction, hypertension, CPB time, RBC infusion and his- cardiopulmonary bypass. Cardiopulmonary bypass can
tory of cardiac surgery and achieved an AUC of 0.84 [16]. cause ischemia-reperfusion injury and systemic systemic
The incidence of pneumonia in that study was 6.6%. inflammation of cardiomyocytes, resulting in decreased
This study also suggested that cardiopulmonary bypass lung compliance. In addition, blood contact with the duct
lasting more than 2 h was an important cause of post- can also cause inflammation in various parts of the body.
operative POP, which may be related to the damage to Kilic et al. found that patients who received CPB for more
Wang et al. Journal of Cardiothoracic Surgery (2024) 19:309 Page 6 of 10

Table 2 Multivariate analysis of risk factors for pneumonia after cardiac surgery
Coefficients () Exp () 95% CI P value Standard error

Intercept -4.3427 - - - 0.3133

Age > 55 years 0.6024 1.83 1.10-3.12 0.0225 0.2640
Malnutrition 1.3108 3.71 2.10-6.43 <0.0001 0.2553
Diabetes mellitus 0.8455 2.33 1.30-4.08 0.0036 0.2908
CPB > 135 min 1.0309 2.80 1.71-4.68 <0.0001 0.2559
Moderate to severe ARDS 0.5804 1.79 1.12-2.85 0.0148 0.2382
Use of ECMO or IABP or CRRT​ 0.9572 2.60 1.31-5.10 0.0057 0.3459
MV > 20 hours 1.1348 3.11 1.88-5.12 <0.0001 0.2553
CPB CardioPulmonary Bypass, ARDS Acute Respiratory Distress Syndrome, ECMO ExtraCorporeal Membrane Oxygenation, IABP Intra-Aortic Balloon Pump, CRRT​
Continuous Renal Replacement Therapy, MV Mechanical Ventilation

Fig. 1 Nomogram based on independent risk factors

than 100 minutes had a 1.7-fold increased risk of various as endotracheal intubation and duration of mechanical
complications and POP after cardiac surgery [10]. Allou ventilation (MV), but also due to impaired local (alveolar)
et al. also reported a positive correlation between CPB and systemic defenses, as well as other specific and non-
duration and POP after cardiac surgery, and in multivari- specific factors [19]. We also found that use of ECMO or
ate analyses, the risk of POP increased significantly with IABP or CRRT was independent risk factor for POP. Cas-
increasing CPB duration [17]. telli et al. report that patients who require IABP support
Acute respiratory distress syndrome (ARDS) recom- after cardiac surgery are more likely to develop postop-
bines a family of disorders with consequences of pneu- erative pneumonia because low cardiac output is harm-
monia, alveolar damage, and pulmonary edema [18, 19]. ful to the immune system [20]. Bizzarro et al. reported
Regardless of the initial lung injury, patients with ARDS a high prevalence of nosocomial infections of 21% in
are prone to lung infection [18]. While may be due to patients requiring ECMO support, with lung infections
bronchial contamination due to traditional factors such being the most common [21]. In addition, critically ill
Wang et al. Journal of Cardiothoracic Surgery (2024) 19:309 Page 7 of 10

Fig. 2 AUC of the nomogram

Fig. 3 Calibration plot of the nomogram

patients treated with CRRT are at risk of lung infection with postoperative pneumonia may be associated with
[22]. Patients requiring IABP, ECMO, and/or CRRT are concomitant critical illness, prolonged mechanical sup-
usually in a state of low cardiac output leading to require- port, mechanical ventilation (MV) and ICU stay, and
ment of such support devices and the harmful effects on impairment of the immune system that promotes the
the immune system. These increased rates of infection release of inflammatory mediators [20–22]. Extended
Wang et al. Journal of Cardiothoracic Surgery (2024) 19:309 Page 8 of 10

Fig. 4 The curve of the incidence of pneumonia with prolonged hospitalization in patients with malnutrition or not

mechanical ventilation has been reported in many studies pneumonia, with Acinetobacter baumannii having the
to be strongly associated with an increased risk of POP, highest detection rate [26]. Anti-infective therapy should
which is associated with damage to respiratory defense be actively given to reduce the level of inflammatory fac-
mechanisms caused by endotracheal intubation [23]. tors in the blood, improve the patient’s immunity, and
We found that malnutrition was a risk factors affecting reduce the probability of MDRO infection in patients
POP, however, previous studies have rarely looked at the with postoperative pneumonia.
relationship between malnutrition and POP. Malnutri- The nomogram models can be used in the clinic for
tion is a complex organismal state and is recognized as joint diagnosis or prediction of disease risk or progno-
a prognostic risk factor for cardiovascular disease [24]. sis by multiple indicators. In addition, the nomogram
A study published by Chermesh et al. demonstrated a models can provide an accurate digital survival or risk
high risk of complex postoperative course, prolonged probability for each patient, which can assist clini-
ICU stay, and increased 3-year mortality in a popula- cians in decision-making and reflect the idea of indi-
tion of malnourished cardiac patients [25]. These factors vidualized medicine. Previous studies have shown that
contribute to postoperative pneumonia. Timely identi- preoperative respiratory physiotherapy and subglottic
fication of patients with malnutrition or at risk of mal- secretion drainage significantly reduce the incidence
nutrition and active and effective nutritional therapy is of POP, pulmonary atelectasis, and other complica-
important for improving the clinical prognosis of cardiac tions [27–29]. In addition, a growing number of studies
surgery patients. have found that aggressive and systematic preopera-
In addition, multidrug-resistant organisms (MDRO) tive oral care plays an important role in preventing and
have become the major bacterial group responsible for minimizing the development of POP after cardiovascu-
pneumonia infections in postoperative cardiac patients lar surgery, with tooth brushing being one of the most
due to bacterial mutations and overuse of antimicro- acceptable and common measures [30, 31]. However, it
bial drugs. A study showed that postoperative pneumo- may not be appropriate to apply these measures to all
nia caused by multidrug-resistant microorganisms was patients without selection, as some measures are time-
associated with adult patients with renal disease, longer consuming, expensive, and laborious. Therefore, the
intraoperative extracorporeal circulation time, and premise of nomogram application is that there must
postoperative nasogastric tubes, with Gram-negative be clear clinical problems and model construction, and
bacilli being the predominant pathogens in patients with its performance and limitations need to be understood
Wang et al. Journal of Cardiothoracic Surgery (2024) 19:309 Page 9 of 10

before being applied to clinical decision-making. Only Availability of data and materials
The datasets generated and/or analyzed during the current study are not pub‑
in this way can nomogram be better applied to the licly available [some patients did not allow us to publish their medical records]
clinic. but are available from the corresponding author upon reasonable request.

Declarations
Limitations and future directions
Ethics approval and consent to participate
Our study also has some limitations. First, the data for Ethical approval was obtained from Medical Ethics Committee of Affiliated
this study were derived from a single-center database, Nanjing Drum Tower Hospital, Nanjing University Medical College, in accord‑
with a single patient source, a small study sample size, ance with the principles of the Declaration of Helsinki on September 23, 2020
(2020-249-01). All included patients were required to provide written informed
and the specific cardiac surgical procedures. Second, we consent.
lack external validation of the model, which may limit its
generalizability. Third, we established standardized diag- Consent for publication
Not applicable.
nostic criteria for POP before the start of the study, but
there may be some degree of variability and subjectivity Competing interests
in clinical diagnosis. For example, although we strictly The authors declare no competing interests.
abide by the principles of sputum culture, but because in Author details
the retention process is susceptible to oropharyngeal col- 1
Department of Cardio‑Thoracic Surgery, Nanjing Drum Tower
onization bacteria contamination, its quality is difficult to Hospital,Affiliated Clinical College of Xuzhou Medical University, Nan‑
jing 210008, Jiangsu, China. 2 Department of Cardio‑Thoracic Surgery, Nanjing
ensure, false positive rate, may be poorly in line with the Drum Tower Hospital, Chinese Academy of Medical Science & Peking Union
diagnosis of clinical infections, and cannot be 100% guar- Medical College, Nanjing 210008, Jiangsu, China. 3 Department of Cardio‑Tho‑
antee of sputum culture positive. This can lead to over- racic Surgery, Nanjing Drum Tower Hospital, Affiliated Hospital of Medical
School, Nanjing University, Nanjing 210008, Jiangsu, China.
estimation or underestimation of the true incidence of
POP. A well-designed prospective study may be needed Received: 23 July 2023 Accepted: 25 May 2024
in the future to obtain more accurate estimates of the
incidence of POP. Future studies should focus on improv-
ing the generalization and practicability of our results.
References
1. Kinlin LM, Kirchner C, Zhang H, et al. Derivation and validation of a clinical
Conclusion prediction rule for nosocomial pneumonia after coronary artery bypass
graft surgery. Clin Infect Dis. 2010;50(4):493–501. https://​doi.​org/​10.​1086/​
Three preoperative indicators (age, preoperative mal- 649925.
nutrition, diabetes mellitus), one intraoperative indica- 2. Horvath KA, Acker MA, Chang H, et al. Blood transfusion and infection
tor (CPB > 135 min) and three postoperative indicators after cardiac surgery. Ann Thorac Surg. 2013;95(6):2194–201. https://​doi.​
org/​10.​1016/j.​athor​acsur.​2012.​11.​078.
(moderate to severe ARDS, use of ECMO or IABP or 3. Zardi EM, Chello M, Zardi DM, et al. Nosocomial extracardiac infections
CRRT, MV> 20 hours) were identified as independent after cardiac surgery [J]. Curr Infect Dis Rep. 2022;24(11):159–71. https://​
risk factors by multivariate logistic regression analy- doi.​org/​10.​1007/​s11908-​022-​00787-0.
4. Strobel RJ, Liang Q, Zhang M, et al. A Preoperative Risk Model for Post‑
sis. A facile nomogram for predicting pneumonia after operative Pneumonia After Coronary Artery Bypass Grafting. Ann Thorac
cardiac surgery was constructed and well validated. A Surg. 2016;102(4):1213–9. https://​doi.​org/​10.​1016/j.​athor​acsur.​2016.​03.​
nomogram performed well in terms of calibration and 074.
5. Gelijns AC, Moskowitz AJ, Acker MA, et al. Management practices and
discrimination, and may have good clinical usefulness. major infections after cardiac surgery. J Am Coll Cardiol. 2014;64(4):372–
Through individualized risk assessment and identifi- 81. https://​doi.​org/​10.​1016/j.​jacc.​2014.​04.​052.
cation of high-risk populations, nomograms can help 6. Chen LF, Arduino JM, Sheng S, et al. Epidemiology and outcome of
major postoperative infections following cardiac surgery: risk factors and
clinicians improve clinical decision-making and help impact of pathogen type. Am J Infect Control. 2012;40(10):963–8. https://​
patients make informed decisions. doi.​org/​10.​1016/j.​ajic.​2012.​01.​012.
7. Massart N, Mansour A, Ross JT, et al. Mortality due to hospital-
Acknowledgements acquired infection after cardiac surgery. J Thorac Cardiovasc Surg.
None. 2022;163(6):2131–40 e3. https://​doi.​org/​10.​1016/j.​jtcvs.​2020.​08.​094.
8. Wang DS, Huang XF, Wang HF, et al. Clinical risk score for postopera‑
Authors’ contributions tive pneumonia following heart valve surgery. Chin Med J (Engl).
KW and HTZ carried out the study, participated in the statistical analysis, and 2021;134(20):2447–56. https://​doi.​org/​10.​1097/​CM9.​00000​00000​
drafted the manuscript. FDF, JP, and TP participated in the statistical analysis. 001715.
DJW conceived of the study, participated in its design and coordination, and 9. Wang D, Huang X, Wang H, et al. Risk factors for postoperative
helped draft the manuscript. All authors have read and approved the final pneumonia after cardiac surgery: a prediction model. J Thorac Dis.
manuscript. 2021;13(4):2351–62. https://​doi.​org/​10.​21037/​jtd-​20-​3586.
10. Kilic A, Ohkuma R, Grimm JC, et al. A novel score to estimate the
Funding risk of pneumonia after cardiac surgery. J Thorac Cardiovasc Surg.
None. 2016;151(5):1415–20. https://​doi.​org/​10.​1016/j.​jtcvs.​2015.​12.​049.
Wang et al. Journal of Cardiothoracic Surgery (2024) 19:309 Page 10 of 10

11. Bonell A, Azarrafiy R, Huong VTL, et al. A Systematic Review and Meta- 31. Nicolosi LN, Del Carmen Rubio M, Martinez CD, Gonzalez NN, Cruz ME.
analysis of Ventilator-associated Pneumonia in Adults in Asia: An Analysis Effect of oral hygiene and 012% chlorhexidine gluconate oral rinse in
of National Income Level on Incidence and Etiology. Clin Infect Dis. preventing ventilator-associated pneumonia after cardiovascular surgery.
2019;68(3):511–8. https://​doi.​org/​10.​1093/​cid/​ciy543. Resp Care. 2014;59:504–9. https://​doi.​org/​10.​4187/​respc​are.​02666.
12. Liu J, Liu J, Wang J, et al. Prevalence and impact of malnutrition on
readmission among hospitalized patients with heart failure in China. ESC
Heart Fail. 2022;9(6):4271–9. https://​doi.​org/​10.​1002/​ehf2.​14152. Publisher’s Note
13. Zhang HT, Han XK, Wang CS, et al. Diagnosis of infection after car‑ Springer Nature remains neutral with regard to jurisdictional claims in pub‑
diac surgery (DICS): a study protocol for developing and validating lished maps and institutional affiliations.
a prediction model in prospective observational study. BMJ Open.
2021;11(9):e048310. https://​doi.​org/​10.​1136/​bmjop​en-​2020-​048310.
14. American Thoracic Society. Infectious Diseases Society of A. Guidelines
for the management of adults with hospital-acquired, ventilator-associ‑
ated, and healthcare-associated pneumonia. Am J Respir Crit Care Med.
2005;171(4):388–416. https://​doi.​org/​10.​1164/​rccm.​200405-​644ST.
15. 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. Clin Infect Dis. 2016;63(5):e61–111. https://​
doi.​org/​10.​1093/​cid/​ciw353.
16. Wang D, Chen X, Wu J, et al. Development and Validation of Nomogram
Models for Postoperative Pneumonia in Adult Patients Undergoing Elec‑
tive Cardiac Surgery. Front Cardiovasc Med. 2021;8:750828. https://​doi.​
org/​10.​3389/​fcvm.​2021.​75082​8[publi​shedO​nline​First:​20211​011].
17. Allou N, Bronchard R, Guglielminotti J, et al. Risk factors for postoperative
pneumonia after cardiac surgery and development of a preoperative risk
score* [J]. Crit Care Med. 2014;42(5):1150–6.
18. Luyt CE, et al. Pulmonary infections complicating ARDS. Intensive Care
Med. 2020;46(12):2168–83. https://​doi.​org/​10.​1007/​s00134-​020-​06292-z.
19. Thompson BT, Chambers RC, Liu KD. Acute respiratory distress syndrome.
N Engl J Med. 2017;377(6):562–72. https://​doi.​org/​10.​1056/​NEJMr​a1608​
077.
20. Castelli P, et al. Intra-aortic balloon counterpulsation: outcome in cardiac
surgical patients. J Cardiothorac Vasc Anesth. 2001;15(6):700–3. https://​
doi.​org/​10.​1053/​jcan.​2001.​28312.
21. Bizzarro MJ, et al. Infections acquired during extracorporeal membrane
oxygenation in neonates, children, and adults. Pediatr Crit Care Med.
2011;12(3):277–81. https://​doi.​org/​10.​1097/​PCC.​0b013​e3181​e28894.
22. Zuo M, et al. Characteristics and factors associated with nosocomial
pneumonia among patients undergoing continuous renal replacement
therapy (CRRT): A case-control study. Int J Infect Dis. 2018;68:115–21.
https://​doi.​org/​10.​1016/j.​ijid.​2018.​01.​008.
23. Papazian L, Klompas M, Luyt CE. Ventilator-associated pneumonia in
adults: a narrative review. Intensive Care Med. 2020;46(5):888–906.
https://​doi.​org/​10.​1007/​s00134-​020-​05980-0.
24. Hill A, et al. Current Evidence about Nutrition Support in Cardiac Surgery
Patients-What Do We Know? Nutrients. 2018;10(5):597. https://​doi.​org/​10.​
3390/​nu100​50597.
25. Chermesh I, Hajos J, Mashiach T, Bozhko M, Shani L, Nir R-R, et al.
Malnutrition in cardiac surgery: food for thought. Eur J Prev Cardiol.
2014;21(4):475–83. https://​doi.​org/​10.​1177/​20474​87312​452969.
26. Wang M, Xu X, Wu S, et al. Risk factors for ventilator-associated pneumo‑
nia due to multi-drug resistant organisms after cardiac surgery in adults
[J]. BMC Cardiovasc Disord. 2022;22(1):465.
27. Katsura M, Kuriyama A, Takeshima T, Fukuhara S, Furukawa TA. Preopera‑
tive inspiratory muscle training for postoperative pulmonary compli‑
cations in adults undergoing cardiac and major abdominal surgery.
Cochrane Database Syst Rev. 2015;10:CD01035. https://​doi.​org/​10.​1002/​
14651​858.​CD010​356.​pub2.
28. Hudson JKC, McDonald BJ, MacDonald JC, Ruel MA, Hudson CCC. Impact
of subglottic suctioning on the incidence of pneumonia after cardiac
surgery: a retrospective observational study. J Cardiothor Vasc An.
2015;29:59–63. https://​doi.​org/​10.​1053/j.​jvca.​2014.​04.​026.
29. Hulzebos EHJ, Smit Y, Helders PPJM, van Meeteren NLU. Preoperative
physical therapy for elective cardiac surgery patients. Cochrane Db Syst
Rev. 2012;11:D10118. https://​doi.​org/​10.​1002/​14651​858.​CD010​118.​pub2.
30. Akutsu Y, Matsubara H, Shuto K, Shiratori T, Uesato M, Miyazawa Y, et al.
Pre-operative dental brushing can reduce the risk of postoperative pneu‑
monia in esophageal cancer patients. Surgery. 2010;147:497–502. https://​
doi.​org/​10.​1016/j.​surg.​2009.​10.​048.