Chronic subclinical inflammation has been implicated in the development of a variety of cardiovascular pathologies.Reference Wrigley, Lip and Shantsila1 Early reports have established that pro-inflammatory biomarkers are increased in patients with cardiac disease such as ischemic and non-ischemic cardiomyopathy.Reference Weir, Miller and Petrie2,Reference Mann3 Pro-inflammatory cytokines, C-reactive protein, and erythrocyte sedimentation rate have also been linked to myocardial migration of monocytes, fibrosis, and ventricular remodelling.Reference Mann3,Reference Fertin, Dubois, Belliard, Amouyel, Pinet and Bauters4 Similarly, known markers of inflammation such as leukocytosis and C-reactive protein have been shown to be predictive of outcomes.Reference Madjid, Awan, Willerson and Casscells5–Reference Pearson, Mensah and Alexander8 Recently, among all markers of inflammation, the neutrophil–lymphocyte ratio has received particular attention as a prognostic marker in adults with cardiovascular disease. Patients with coronary artery disease (both stable angina and acute coronary syndrome), myocardial infarction, and those undergoing coronary interventions (both surgical and percutaneous) have been shown to exhibit a higher risk of adverse outcomes in the setting of baseline neutrophil–lymphocyte elevation.Reference Papa, Emdin, Passino, Michelassi, Battaglia and Cocci9–Reference Gurm, Bhatt and Lincoff13 Data in the population of children undergoing congenital heart surgery are lacking. Therefore, the objective of this study was to evaluate the association of pre- and post-operative neutrophil–lymphocyte ratio with post-operative outcomes in children after congenital heart surgery. We hypothesised that elevated neutrophil–lymphocyte ratio is associated with adverse outcomes such as severity of low cardiac output syndrome and vasoactive infusion use.
Materials and methods
We performed a post hoc analysis of data collected as part of a prospective observational cohort of children evaluated previously for glucocorticoid receptor levels after heart surgery.Reference Flores, Cooper and Opoka14 The initial cohort consisted of children (up to 17 years of age) admitted to the cardiac ICU of the Cincinnati Children’s Hospital following congenital heart surgery with cardiopulmonary bypass during a 7-month period (1 July, 2015 to 31 January, 2016). Patients who did not require cardiopulmonary bypass were excluded. Clinical and other laboratory data were captured daily until hospital discharge or 28 days (whichever occurred first). All patients were administered 30 mg methylprednisolone as per standard practice at the onset of cardiopulmonary bypass. We sought to evaluate the association of pre- and post-operative neutrophil–lymphocyte ratio with post-operative outcomes such as low cardiac output and vasoactive medication use.
The primary endpoint was low cardiac output syndrome defined based on the number of vasoactive infusions (1, 2, >2) and level of blood lactate (<2, 2–5, >5 in mmol/l). Patients were categorised into mild low cardiac output group if they only required one vasoactive infusion and had blood lactate <2; moderate – two vasoactive infusions and had blood lactate 2–5; severe – three or more vasoactive agents or blood lactate >5; mechanical circulatory support – extracorporeal membrane oxygenation or ventricular assist devices; and finally, death. We used blood lactate and number of vasoactive agents to define low cardiac output because both these endpoints have been predictive of worse clinical outcomes in the population of children with CHD.Reference Hannan, Ybarra, White, Ojito, Rossi and Burke15,Reference Gaies, Gurney and Yen16
The secondary endpoint was the time to inotrope free defined as the number of days until the patient was weaned off vasoactive infusions.Reference Flores, Cooper and Opoka14 Post-operative day 1 was defined as the time from return from the operating room until midnight of the same day. Post-operative days 2 and 3 were defined from midnight to midnight of the following 2 days. The study received approval by the Institutional Review Board of Cincinnati Children’s Hospital. The requirement for informed consent was waived given its retrospective nature.
Statistical analysis
Summary statistics were presented as median and interquartile range for continuous variables and as counts and percentages for categorical variables. The independent variable neutrophil–lymphocyte ratio was examined pre-operatively and post-operatively in relation to each outcome. For analyses, low cardiac output syndrome scores were examined as three categories: mild, moderate, and severe (including mechanical circulatory support and death), given the low observed incidence of mechanical support and death.Reference Flores, Cooper and Opoka14 Cumulative odds for severe and moderate/low cardiac output versus mild low cardiac output were estimated using proportional odds models, and the results were presented as odds ratios and 95% confidence interval. The time to inotrope free was examined as time to event analyses using Cox regression models, and the results were reported as hazard ratio and 95% confidence interval. In this study, shorter time to inotrope free was favourable outcome in contrast to usual survival analyses, whereas longer time is the favourable outcome. Therefore, in this occasion, hazard ratio more than one was positive outcome indicating shorter time to inotrope free. Vasoactive-inotrope scores were examined using linear regression analyses, and the results were presented as beta estimates and standard error. Significant covariates from the univariate analyses for each outcome were adjusted for in the respective multi-variable models. All analyses were performed as two-sided test using SAS 9.4 (SAS Institute Inc., Cary, North Carolina, United States of America) and p 0.05 was considered significant.
Results
The cohort consisted of 83 children, but only 47 had complete blood count with differential available for neutrophil–lymphocyte ratio analysis. Most study patients were infants (80%) with a median age of 4.1 months (interquartile range 0.2–7.6) and a median weight 5.1 kg (interquartile range 3.5–7.2). The most frequent diagnoses were conotruncal abnormalities (36%) and left-sided obstructed lesions (28%). The demographic of the cohort analysed is provided in Table 1. Three patients suffered mortality and two required mechanical circulatory support. Higher pre-operative neutrophil–lymphocyte ratio was associated with higher cumulative odds of severe and moderate versus mild low cardiac output syndrome for the first post-operative day (odds ratio 2.86 (95% confidence interval 1.18–6.93); p = 0.02), after adjusting for covariates. The results are schematically presented in Fig 1. Patients with increased pre-operative neutrophil–lymphocyte ratio demonstrated a trend of longer time to inotrope free (hazard ratio 0.88; 95% confidence interval 0.66–1.17; p = 0.37), although this was not statistically significant (Table 2). Similarly, increased pre-operative neutrophil–lymphocyte ratio showed a positive association on univariate analysis with higher vasoactive-inotrope scores on post-operative day 1, but this association did not reach significance level (β estimate ± standard error (0.11 ± 0.11); p = 0.30) (Table 3). In the contrary, post-operative neutrophil–lymphocyte ratio showed no association with time to inotrope free, low cardiac output score, and vasoactive-inotrope score.
Table 1. Patient characteristics
Figure 1. Association between neutrophil–lymphocyte ratio and low cardiac output syndrome. Adjusted for cardiopulmonary bypass time, aortic cross-clamp time, Risk Adjustment for Congenital Heart Surgery-1 scores, and post-op hydrocortisone use (yes/no). Proportional odds analyses performed and cumulative odds for severe and moderate versus mild low cardiac output syndrome examined. NLR = neutrophil–lymphocyte ratio.
Table 2. Association between neutrophil–lymphocyte ratio and time to inotrope free
HR = hazard ratio; LCL = lower control limit; NLR = neutrophil–lymphocyte ratio; UCL = upper control limit.
1 Adjusted for age, weight, cardio pulmonary bypass time, aortic cross clamp time, Risk Adjustment for Congenital Heart Surgery-1 scores, and post-operative hydrocortisone use (yes/no)
2 Cox regression analyses performed and HR and confidence intervals presented. HR > 1 indicates shorter Time to Inotrope Free
Table 3. Association between neutrophil–lymphocyte ratio and vasoactive-inotrope score1
NLR = neutrophil–lymphocyte ratio; POD = post-operative day; SE = standard error; VIS = vasoactive-inotrope score.
1 Linear regression analyses performed
2 Adjusted for cardiopulmonary bypass time, aortic cross clamp time, Risk Adjustment for Congenital Heart Surgery-1 score and post-operative hydrocortisone use (yes/no)
Discussion
Our study documented a significant relationship between pre-operative neutrophil–lymphocyte ratio and low cardiac output syndrome after cardiac surgery in children. In particular, patients with elevated pre-operative neutrophil–lymphocyte ratio are more likely to suffer moderate or severe low cardiac output syndrome during the first 12 post-operative hours. This is the first study to document the prognostic significance of neutrophil–lymphocyte ratio in children after cardiac surgery.
Similar findings, however, were recently reported in adults. In a cohort of 3027 patients after surgery with cardiopulmonary bypass for coronary artery or valvar disease, Silberman et al.Reference Silberman, Abu-Yunis and Tauber17 reported that an elevated neutrophil–lymphocyte ratio was an independent predictor of low cardiac output syndrome (hazard ratio 1.54; 95% confidence interval 1.23–1.93; p = 0.003) and other morbidities such as prolonged ventilation and pleural effusion. Interestingly enough, this study indicated that an elevated neutrophil ratio retained its prognostic significance even in the setting of a normal neutrophil count. More importantly, elevated neutrophil–lymphocyte ratio was an independent risk factor for early as well as late operative mortality in this cohort. Our study was not adequately powered to detect differences in mortality, given the small sample size and the low incidence of operative mortality.
The literature linking the neutrophil–lymphocyte ratio with outcomes in patients with cardiac disease continues to grow. In adults with heart failure, the neutrophil–lymphocyte ratio has been associated with outcomes irrespective of disease severity. In a recent cohort of 273 adults who required placement of a left ventricular assist device, the neutrophil–lymphocyte ratio was found to be independent predictor of post-operative right ventricular failure and mortality.Reference Yost, Joseph, Tatooles and Bhat18 Similarly, in medically treated adults with acute heart failure, Uthamalingham et al.Reference Uthamalingam, Patvardhan and Subramanian19 demonstrated that the neutrophil–lymphocyte ratio (even in the setting of normal neutrophil count) was independently associated with in-hospital mortality, regardless of baseline left ventricular function. Similarly, Polat et al.Reference Polat, Yildiz and Bilik20 showed that a decreased lymphocyte count correlated with mortality in a cohort of 119 hospitalised adults with heart failure and reduced ejection fraction. Recently, in a large cohort of 1923 hospitalised adults with heart failure with either preserved or reduced ejection fraction, Huang et al.Reference Huang, Cheng and Huang21 demonstrated that a number of haemographic indices (including neutrophil–lymphocyte ratio) were independent predictors of mortality. Increased neutrophil as well as reduced lymphocyte count has been associated with sudden cardiac death in patients with left ventricular dysfunction.Reference Cooper, Exner, Waclawiw and Domanski22,Reference Huehnergarth, Mozaffarian and Sullivan23 Finally, in a very large adult population study, Shah et al.Reference Shah, Denaxas, Nicholas, Hingorani and Hemingway24 documented a higher cumulative incidence of cardiovascular disease such as coronary death, heart failure, non-fatal myocardial infarct, and peripheral arterial disease in patients with neutrophil counts in the upper range of normal. Interestingly, these associations appeared stronger in patients in stable clinical condition and when means of two counts were used, probably attesting to the pathophysiologic relevance of chronic inflammation.
Our study does not provide data to explain the mechanism linking the pre-operative elevation of neutrophil–lymphocyte ratio and the severity of low cardiac output syndrome or other potential adverse outcomes. We are, therefore, unable to answer the question as to whether the additional risk is conferred by the circulating neutrophils per se or a potential chronic inflammatory state of which the neutrophil count is just a marker. We can only speculate based on known pathophysiologic mechanisms and evidence produced in previous studies. Chronic inflammation has been previously implicated in the pathophysiology of heart failure and possibly other types of cardiovascular disease.Reference Mann3,Reference Shah, Denaxas, Nicholas, Hingorani and Hemingway24,Reference Briasoulis, Androulakis, Christophides and Tousoulis25 Pro-inflammatory cytokines are part of the stress response (alongside neuro-hormonal activation) that aims to restore homeostasis.Reference Briasoulis, Androulakis, Christophides and Tousoulis25 Neutrophils participate in initiating the inflammatory response and have been recently recognised as decision-shapers during innate and adaptive immune responses which are necessary for healing after tissue injury.Reference Frangogiannis26 They secrete proteolytic enzymes that clear the necrotic cells and extracellular matrix promoting tissue repair.Reference Frangogiannis27 Experimental data have demonstrated the rapid development of heart failure after myocardial infarction in neutrophil-depleted animals.Reference Horckmans, Ring and Duchene28 On the other hand, neutrophils amplify inflammatory response.Reference Brinkmann, Reichard and Goosmann29,Reference Dalli, Montero-Melendez and Norling30 Over activation of the inflammatory processes can, potentially, become maladaptive and have deleterious results in myocardial function.Reference Briasoulis, Androulakis, Christophides and Tousoulis25 Evidence suggests that it can lead to impaired cardiac metabolism, ventricular dysfunction, negative inotropic effects, and possibly myocardial remodelling. In that light, neutrophils play a key role in inflammatory-mediated tissue injury and have been associated with reperfusion injury, atherosclerotic plague disruption, and hypercoagulability.Reference Soehnlein, Steffens, Hidalgo and Weber31–Reference Segel, Halterman and Lichtman33 It is plausible that the fine balance between reparative tissue healing mediated by a “healthy” inflammatory/stress response and deleterious inflammatory overactivation is linked to the action of both neutrophils, as markers of non-specific inflammation, and lymphocytes, as markers of immunologic stress response.
However, studies with mixed and valvar disease have also supported the prognostic value of neutrophil–lymphocyte ratio, whereas no inflammatory mechanism has been suggested in the pathophysiology of valvar disease.Reference Silberman, Abu-Yunis and Tauber17,Reference Habib, Thawabi and Hawatmeh34 The link remains elusive. These data may strengthen the notion that neutrophil and lymphocyte counts reflect non-specific inflammatory and immune processes which contribute to myocardial dysfunction in the setting of surgical intervention. Nevertheless, more work is clearly needed before we can confidently attest to the potential pathophysiologic connection.
Similarly, we cannot comment on the question that naturally arises, whether prophylactic pre-operative treatment with anti-inflammatory therapies can potentially modify adverse outcomes. No such data are provided by our study or others in the paediatric cardiac population. Immunomodulatory agents have been shown to reduce the level of cytokines in small clinical studies of heart failure patients but failed to demonstrate positive impact on outcomes in double-blind placebo-controlled studies.Reference Mann3,Reference Sliwa, Woodiwiss and Candy35,Reference Krum, Ashton and Reid36 This is a very interesting area that will most likely receive attention in future research. However, based on the current evidence, we can only advocate for the use of the lymphocyte–neutrophil ratio in pre-operative risk assessment and stratification.
Our study has several limitations. Firstly, the retrospective collection of neutrophil–lymphocyte ratio data may have introduced selection bias. Similar to any observational study, causation cannot be implied due to the possibility of cofounding. However, this is the first study to establish the association of neutrophil–lymphocyte ratio with post-operative outcomes after cardiac surgery in children. The findings are in keeping with the literature in adults with cardiovascular disease and set the ground for larger prospective confirmatory studies. The retrospective nature of our review also precluded us from studying other well-established markers of inflammation such as C-reactive protein, procalcitonin, and erythrocyte sedimentation ratio that would have potentially shed more light into the inflammatory status of the cohort. These tests are not routinely included in the pre-operative assessment of paediatric cardiac surgery patients.
Secondly, we are limited from the small sample size because of retrospective collection of neutrophil–lymphocyte ratio data for a prospective cohort. However, despite the small sample size, we found significant associations. It is likely that these associations will emerge stronger in larger studies. Another limitation related to our small sample size is our inability to evaluate the various subgroups that may constitute the group of patients with elevated neutrophil–lymphocyte ratio: elevated neutrophil count versus low lymphocyte count versus normal neutrophil and lymphocyte count but abnormal ratio. As already mentioned, adult studies suggest that elevated neutrophil–lymphocyte ratio in itself has prognostic significance even in the setting of normal neutrophil count.Reference Silberman, Abu-Yunis and Tauber17
Finally, an important limitation is the heterogeneity of our study subjects. We included patients undergoing different procedures with potentially different anatomic and physiologic baseline characteristics with regards to pulmonary/systemic blood flow balance, ventricular function, and degree of baseline heart failure. However, large adult studies after cardiac surgery have indicated that although elevated neutrophil–lymphocyte ratio was associated with more severe baseline morbidity, it remained a predictor of adverse effects independent of baseline characteristics.Reference Silberman, Abu-Yunis and Tauber17–Reference Huang, Cheng and Huang21 Large multi-center studies in the paediatric cardiac population are required to address the issue of heterogeneous baseline characteristics and confirm generalisability of our findings.
The neutrophil–lymphocyte ratio prior to cardiac surgery in children is predictive of severity of low cardiac output syndrome in the immediate post-operative period. It is an easily obtained and inexpensive marker that can offer an additional means of risk stratification in this vulnerable population. Further studies in large multi-center cohorts are required to confirm validity and generalisability of these preliminary findings.
Acknowledgements
None.
Financial Support
This research received no specific grant from any funding agency commercial or not-for-profit sectors.
Conflicts of Interest
None.
Ethical Standards
All study procedures complied with the ethical standards of the Helsinki Declaration and has been approved by Institutional Research Board of Cincinnati Children’s Medical Centre.
