Heart rate recovery and morbidity after noncardiac surgery: Planned secondary analysis of two prospective, multi-centre, blinded observational studies

PLOS ONE, Aug 2019

Background Impaired cardiac vagal function, quantified preoperatively as slower heart rate recovery (HRR) after exercise, is independently associated with perioperative myocardial injury. Parasympathetic (vagal) dysfunction may also promote (extra-cardiac) multi-organ dysfunction, although perioperative data are lacking. Assuming that cardiac vagal activity, and therefore heart rate recovery response, is a marker of brainstem parasympathetic dysfunction, we hypothesized that impaired HRR would be associated with a higher incidence of morbidity after noncardiac surgery. Methods In two prospective, blinded, observational cohort studies, we established the definition of impaired vagal function in terms of the HRR threshold that is associated with perioperative myocardial injury (HRR ≤ 12 beats min-1 (bpm), 60 seconds after cessation of cardiopulmonary exercise testing. The primary outcome of this secondary analysis was all-cause morbidity three and five days after surgery, defined using the Post-Operative Morbidity Survey. Secondary outcomes of this analysis were type of morbidity and time to become morbidity-free. Logistic regression and Cox regression tested for the association between HRR and morbidity. Results are presented as odds/hazard ratios [OR or HR; (95% confidence intervals). Results 882/1941 (45.4%) patients had HRR≤12bpm. All-cause morbidity within 5 days of surgery was more common in 585/822 (71.2%) patients with HRR≤12bpm, compared to 718/1119 (64.2%) patients with HRR>12bpm (OR:1.38 (1.14–1.67); p = 0.001). HRR≤12bpm was associated with more frequent episodes of pulmonary (OR:1.31 (1.05–1.62);p = 0.02)), infective (OR:1.38 (1.10–1.72); p = 0.006), renal (OR:1.91 (1.30–2.79); p = 0.02)), cardiovascular (OR:1.39 (1.15–1.69); p<0.001)), neurological (OR:1.73 (1.11–2.70); p = 0.02)) and pain morbidity (OR:1.38 (1.14–1.68); p = 0.001) within 5 days of surgery. Conclusions Multi-organ dysfunction is more common in surgical patients with cardiac vagal dysfunction, defined as HRR ≤ 12 bpm after preoperative cardiopulmonary exercise testing. Clinical trial registry ISRCTN88456378.

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0221277&type=printable

Heart rate recovery and morbidity after noncardiac surgery: Planned secondary analysis of two prospective, multi-centre, blinded observational studies

August Heart rate recovery and morbidity after noncardiac surgery: Planned secondary analysis of two prospective, multi-centre, blinded observational studies Gareth L. AcklandID 0 2 3 Tom E. F. AbbottID 0 2 3 Gary Minto 1 2 3 Martin Clark 2 3 Thomas Owen 2 3 Pradeep Prabhu 2 3 Shaun M. MayID 0 2 3 Joseph A. Reynolds 0 2 3 Brian H. Cuthbertson 2 3 Duminda Wijesundera 2 3 Rupert M. Pearse 0 2 3 for the METS 2 3 POM-HR Study Investigators 2 3 0 William Harvey Research Institute, Queen Mary University of London , London , United Kingdom 1 Department of Anaesthesia, Derriford Hospital, Plymouth Hospitals NHS Trust; Peninsula Schools of Medicine and Dentistry, Plymouth University , Plymouth , United Kingdom , 3 Department of Anaesthesia, Royal Bournemouth Hospital , Bournemouth , United Kingdom , 4 Department of Anaesthesia, Royal Preston Hospital, Lancashire Teaching Hospitals NHS Foundation Trust, Preston, United Kingdom, 5 Department of Anaesthesia, Royal Surrey County Hospital , Guildford , United Kingdom , 6 Department of Critical Care Medicine, Sunnybrook Health Sciences Centre , Toronto, Ontario , Canada , 7 Department of Anesthesia, University of Toronto , Toronto, Ontario , Canada , 8 Li Ka Shing Knowledge Institute, St. Michael's Hospital , Toronto, Ontario , Canada , 9 Department of Anesthesia and Pain Management, Toronto General Hospital , Toronto, Ontario , Canada 2 Editor: Cesario Bianchi, Universidade de Mogi das Cruzes , BRAZIL 3 Funding: The METS Study was funded by peerreviewed grants from the Canadian Institutes of Health Research, Heart and Stroke Foundation of Canada, Ontario Ministry of Health and Long-Term Care, Ontario Ministry of Research and Innovation, National Institute of Academic Anaesthesia, UK Clinical Research Network , Australian and New - Data Availability Statement: All relevant data are within the manuscript and its Supporting Information files. Background Impaired cardiac vagal function, quantified preoperatively as slower heart rate recovery (HRR) after exercise, is independently associated with perioperative myocardial injury. Parasympathetic (vagal) dysfunction may also promote (extra-cardiac) multi-organ dysfunction, although perioperative data are lacking. Assuming that cardiac vagal activity, and therefore heart rate recovery response, is a marker of brainstem parasympathetic dysfunction, we hypothesized that impaired HRR would be associated with a higher incidence of morbidity after noncardiac surgery. Methods In two prospective, blinded, observational cohort studies, we established the definition of impaired vagal function in terms of the HRR threshold that is associated with perioperative myocardial injury (HRR 12 beats min-1 (bpm), 60 seconds after cessation of cardiopulmonary exercise testing. The primary outcome of this secondary analysis was all-cause morbidity three and five days after surgery, defined using the Post-Operative Morbidity Survey. Secondary outcomes of this analysis were type of morbidity and time to become morbidityfree. Logistic regression and Cox regression tested for the association between HRR and morbidity. Results are presented as odds/hazard ratios [OR or HR; (95% confidence intervals). Zealand College of Anaesthetists, and Monash University. POM-HR study was supported by British Journal of Anaesthesia/Royal College of Anaesthetists? Basic Science Career development fellowship [GLA] and the UK Clinical Research Network. TEFA was supported by a Medical Research Council and British Journal of Anaesthesia clinical research training fellowship (grant reference MR/M017974/1); RP is supported by a UK National Institute for Health Research Professorship; GLA is supported by British Journal of Anaesthesia/Royal College of Anaesthetists basic science Career Development award, British Oxygen Company research chair grant in anaesthesia from the Royal College of Anaesthetists and British Heart Foundation Programme Grant (RG/14/4/30736). BHC and DNW are supported in part by Merit Awards from the Department of Anesthesia at the University of Toronto. DNW is supported by a New Investigator Award from the Canadian Institutes of Health Research. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. The specific roles of these authors are articulated in the ?author contributions? section. Competing interests: The METS Study and POMHR funding sources had no role in the design and conduct of these studies; collection, management, analysis, and interpretation of the data; and preparation or approval of the article. RP holds research grants, and has given lectures and/or performed consultancy work for Nestle Health Sciences, BBraun, Medtronic, GlaxoSmithKline and Edwards Lifesciences, and is a member of the Associate editorial board of the British Journal of Anaesthesia; GLA is a member of the editorial advisory board for Intensive Care Medicine Experimental, is an Editor for the British Journal of Anaesthesia and has undertaken consultancy work for GlaxoSmithKline; TEFA is a committee member of the Perioperative Exercise Testing and Training Society. This does not alter our adherence to PLOS ONE policies on sharing data and materials. Results 882/1941 (45.4%) patients had HRR 12bpm. All-cause morbidity within 5 days of surgery was more common in 585/822 (71.2%) patients with HRR 12bpm, compared to 718/1119 (64.2%) patients with HRR>12bpm (OR:1.38 (1.14?1.67); p = 0.001). HRR 12bpm was associated with more frequent episodes of pulmonary (OR:1.31 (1.05?1.62);p = 0.02)), infective (OR:1.38 (1.10?1.72); p = 0.006), renal (OR:1.91 (1.30?2.79); p = 0.02)), cardiovascular (OR:1.39 (1.15?1.69); p<0.001)), neurological (OR:1.73 (1.11?2.70); p = 0.02)) and pain morbidity (OR:1.38 (1.14?1.68); p = 0.001) within 5 days of surgery. Conclusions Clinical trial registry ISRCTN88456378. Multi-organ dysfunction is more common in surgical patients with cardiac vagal dysfunction, defined as HRR 12 bpm after preoperative cardiopulmonary exercise testing. Introduction Reduced efferent vagal neural activity (hereafter, vagal dysfunction) is a common feature of injury and systemic inflammation.[ 1, 2 ] Experimental medicine studies have demonstrated that parasympathetic neurotransmitters released by vagal nerve activity confer organ protection,[3] in part through maintaining exercise capacity[ 4 ] and/or limiting systemic inflammation.[ 5 ] Preservation, or augmentation, of efferent vagal nerve activity reduces cardiac,[ 6 ] pulmonary[ 7 ] and renal injury,[ 8, 9 ] as well as enhancing endogenous analgesic mechanisms [10] and promoting gastrointestinal recovery after surgery.[ 11 ] Slower heart rate recovery (HRR) after exercise is associated with an increased risk of perioperative myocardial injury after noncardiac surgery.[ 12 ] Parasympathetic vagal innervation of the heart can be quantified by heart rate recovery fol lowing peak exercise [ 13?15 ]; heart rate recovery is independent of exercise workload.[ 16 ] Vagal reactivation is the primary mechanism underlying deceleration of heart rate after exercise, as demonstrated by its blockade in humans by atropine.[ 17 ] Athletes exhibit accelerated, vagally mediated heart rate recovery after exercise, in contrast to the blunted response observed in patients with heart failure.[ 17 ] Heart rate control is attributable to neuronal substrate within the nucleus ambiguus. [ 18 ] The other main parasympathetic brainstem component, the dorsal vagal motor nucleus, also regulates cardiac ventricular function, as well as innervating multiple other organs.[ 18 ] Developmental ontologic studies show that neurons from the DVMN complex give rise to neurons comprising the nucleus ambiguous.[ 18 ] Therefore, dysfunction in nucleus ambiguous neurons, is likely to be mirrored in other vagal neurons. These data suggest a mechanistic role for cardiac vagal dysfunction in promoting perioperative myocardial injury. Whether cardiac vagal impairment reflects broader parasympathetic impairment that may promote perioperative organ dysfunction has not been systematically explored. Here, we hypothesized that impaired HRR, a physiologic marker of impaired cardiac parasympathetic activity, was associated with increased postoperative morbidity. 2 / 15 Methods Study design and setting We undertook a prespecified secondary analysis of two international, multi-centre, prospective observational studies utilising cardiopulmonary exercise testing. In both studies, morbidity data were prospectively collected at the same timepoints after surgery, with both patients and clinicians blinded to heart rate recovery. Using these similar datasets, we assessed whether impaired vagal activity was associated with excess postoperative morbidity after noncardiac surgery. Research ethics committees reviewed both studies, which were conducted in accordance with the principles of the Declaration of Helsinki and the Research Governance Framework. The Post Operative Morbidity-Heart Rate recovery (POM-HR) study was approved by UK Medical Research Ethical Committee (MREC):12/LO/0453 [approved 30/5/2012]; ISRCTN88456378). The Measurement of Exercise Tolerance before Surgery (METS) study was approved by UK MREC 13/LO/0135 [approved 15 February 2013]. The POM-HR study prospectively recruited high-risk patients (>40y) in 5 UK centres from to 1/11/2012 to 17/1/2015. Inclusion criteria were: (1) referral for cardiopulmonary exercise testing from the patients surgical and/or anaesthesia preassessment clinic; (2) surgery of duration >2 h; (3) at higher risk of complications after surgery, as estimated by their referring anaesthesiologist and/or surgical service. Patients were excluded if they had ATS-defined contraindications to cardiopulmonary exercise testing. The primary outcome in POM-HR was any postoperative complication as defined by the Postoperative Morbidity Survey (POMS)22 within five days of surgery. The METS observational study was not registered; the protocol and methods were published prior to completion of the study. [19, 20] METS was undertaken at 25 hospitals in Canada, UK, Australia, and New Zealand, recruiting patients from March 1, 2013 to March 25, 2016. Inclusion criteria were: (1) participants >40y (2) elective noncardiac surgery under general or regional anaesthesia (or both) with a minimum of one overnight hospital stay (3) 1 risk factor for cardiac complications (coronary artery disease, cardiac failure, cerebrovascular disease, diabetes mellitus, chronic renal failure, peripheral arterial disease, hypertension, a history of tobacco smoking within 12 months of surgery, or >70y). Exclusion criteria were: endovascular surgery; insufficient time for cardiopulmonary exercise testing (CPET) before surgery; implantable cardioverter?defibrillator; pregnancy; previous enrolment in the study; severe hypertension (>180/100 mm Hg) and/or other American Thoracic Society-defined contraindications to undertaking CPET. The primary outcome in METS was death or myocardial infarction within 30 days after surgery. The secondary outcomes included complications after surgery, as defined by POMS and Clavien-Dindo grading. Cardiopulmonary exercise testing (CPET) Participants undertook CPET on an electronic cycle ergometer to maximal tolerance, having continued their normal cardiovascular medications up to and including the day of the test. [ 21 ] Continuous 12-lead electrocardiogram was recorded. Resting heart rate was recorded before unloaded pedaling in the sitting position. Equipment was calibrated before each test using standard reference gases. Continuous breath-by-breath gas exchange analysis was performed. All patients were instructed to continue cycling until symptom-limited fatigue occurred. After peak effort was reached, workload was reduced to 20W and the participant continued to pedal for five minutes in order to warm-down. Investigators at each site interpreted each CPET and collected a standardised data set (Supplementary data). Clinicians at 3 / 15 each site were blinded to the results of cardiopulmonary exercise testing, except where there was a safety concern according to pre-defined criteria.[19] We calculated heart rate recovery by subtracting heart rate 1 minute after the end of exercise from heart rate at peak exercise. [ 22 ] Personnel and patients were masked to heart rate recovery data; none of the software used in each centre automatically provided these data. Heart rate recovery data were not provided in reports to clinical teams; therefore these data had no influence on subsequent clinical care. Perioperative management Patients were cared for by the normal attending clinicians, who were blinded to HRR results. All hospitals that contributed patients partake in enhanced recovery programs for the types of surgery involved in this observational study. Surgery and anesthesia were conducted by specialist staff. Perioperative care conformed with local clinical guidelines and was not standardised. Exposure of interest The exposure of interest was heart rate recovery, for which we classified HRR as normal or impaired based on values calculated at 1 minute after the end of peak exercise 12 beats.min-1. Previous exercise studies that have enrolled >20000 patients in the general population show that HRR 12 beats.min-1 one minute after cessation of exercise is independently associated with increased mortality. [ 13, 23 ] Outcomes The primary outcome was all-cause postoperative morbidity, assessed using the Post Operative Morbidity Survey (POMS; S1 Table), which was collected prospectively within 5 days of surgery.[2] Secondary outcomes were type of morbidity (as defined by POMS), time to become morbidity-free and length of hospital stay. Sensitivity analysis To examine whether cardiac vagal dysfunction is associated with outcomes after surgery inde pendently of subclinical moderate-severe heart failure, [ 24 ] we repeated the primary analysis to assess whether the presence of delayed HRR remained associated with the primary outcome in the presence or absence of VO2 peak 14 ml/kg/min and/or VE/VCO2 at the anaerobic threshold 34. Preoperative use of beta-blockade, which does not impact negatively on HRR in the cardiac failure population, [ 24 ] [ 25 ] was also subjected to a similar sensitivity analysis. Statistical analysis Manual and automated validation checks of data were performed both centrally and through source data verification. Descriptive categorical data are summarized as counts (percentage). Descriptive continuous data are presented as mean (95% confidence intervals) and analysed using ANCOVA (controlling for age), with post-hoc Tukey Kramer tests to identify within and between factor differences. We present participants? characteristics for the whole cohort and stratified by HRR or >12 beats.min-1. The primary (categorical) outcome was analysed using Fishers exact test for trend. Second ary outcomes were analysed using Fisher?s exact test (type/severity of morbidity) and Cox regression analysis (time-to-become morbidity free), taking into account the following independent variables: age, body-mass index, gender, surgery type (intra-abdominal, orthopaedic, urology/gynaecology, vascular, others), diabetes mellitus, preoperative cardiovascular disease 4 / 15 (ischaemia/heart failure/dysrhythmias), resting heart rate and HRR 12bpm. For log-rank analysis of length of stay, patients who died were right-censored as the largest length of stay. P<0.05 was considered significant. All statistical analyses were undertaken using NCSS 11 (Kaysville, UT, USA). Sample size calculation Sample size was calculated to detect differences in all-cause morbidity on postoperative day 5, assuming that, overall, up to 40% of patients undergoing major surgical procedures may sustain morbidity at this timepoint.[ 26, 27 ] On the basis that a 15% relative risk reduction in allcause morbidity by postoperative day 5 would be of clinical significance, with power of 90%, at least 1920 patients would be required to detect a 15% relative risk reduction in postoperative morbidity comparing patients with HRR 12 beats min-1 versus patients with preserved HRR (? = 0.05). Results Patient characteristics 1741 patients were recruited into the METS study between 1st March 2013 and 25th March 2016. 840 patients were recruited into the POM-HR study between 1st May 2012 and 31st March 2015. 1941 cases were analysed after cases with missing data were excluded; 882/1941 (45.4%) had HRR 12 beats min-1 (Fig 1). Mean resting heart rate was 6 (95%CI:4?7) beats minute-1 higher in patients with HRR 12 beats min-1 (p<0.001, by ANCOVA controlling for age; Fig 2A). Peak heart rate during exercise was 12 (95%CI:10?14) beats minute-1 lower in patients with HRR 12 beats min-1 (p<0.001, by ANCOVA controlling for age; Fig 2A). Systolic and diastolic blood pressure at rest were similar (Table 1). Patients with/without delayed heart recovery had similar preoperative characteristics and underwent similar types of surgery (Table 1). Primary outcome: Morbidity within 5 days of surgery All-cause morbidity within five days of surgery was more common in 585/822 (71.2%) patients with HRR 12bpm, compared to 718/1119 (64.2%) patients with preserved HRR (OR:1.29 (1.06?1.58); p = 0.001; Table 2; Fig 2B). Morbidity on postoperative day 3 and 5 is provided in S2 Table. Logistic regression analysis showed that lower HRR was progressively independently associated with higher morbidity rates within 5 days of surgery, independent of resting heart rate (S3 and S4 Tables). Secondary outcomes Types of morbidity. HRR 12 beats.min-1 was associated with more frequent episodes of morbidity on both postoperative days 3 and 5. (S2 Table) Cardiovascular morbidity was more common in patients with HRR 12 beats.min-1 (OR:1.39 (1.15?1.69); p<0.001)), characterised by more episodes of hypotension (OR:1.71 (1.06?2.76); p = 0.03), more frequent intraoperative use of norepinephrine (OR:3.71 (1.54?8.94); p = 0.003) and a higher proportion of arrythmias detected independent of hypotension (4.1% versus 3.2% respectively). Patients with HRR 12 beats.min-1 sustained more pulmonary (OR:1.31 (1.05?1.62); p = 0.02), including pneumonia (OR:2.52 (1.05?6.05)), infection (OR:1.38 (1.10?1.72); p = 0.006), renal (OR:1.91 (1.30?2.79); p = 0.02), neurological (OR:1.73 (1.11?2.70); p = 0.02) and pain morbidity (OR:1.38 (1.14? 1.68); p = 0.001; S2 Table). 5 / 15 Fig 1. Analysis plan. Time to become morbidity-free. Fewer patients with HRR 12 beats.min-1 (237/882 (28.8%)) remained morbidity-free within 5 days of surgery, compared to 401/1119 (35.8%) patients with preserved HRR (OR:1.35 (1.06?1.70); p = 0.01; S5 Table). Surgery for malignancy (risk ratio:1.26 (1.13?1.40); p<0.001) and HRR 12 beats.min-1 (risk ratio:1.11 (1.05?1.18); 6 / 15 Fig 2. Exercise-evoked heart rate dynamics and distribution of heart rate recovery. A. Exercise-evoked changes in heart rate from baseline to peak, and 60s after recovery from the end of exercise in patients with delayed, or normal, heart rate recovery >12beats.min-1. Red colour indicate values below which are associated with increased risk of all-cause mortality obtained in large epidemiological studies.[ 13, 14 ] Median (25-75th centile values shown; p values determined by ANCOVA, controlling for age. B. Distribution of heart 7 / 15 rate recovery (left axis) plotted against all-cause morbidity (right axis) within 5 days of surgery. Red bars indicate values below which are associated with increased risk of all-cause mortality obtained in large epidemiological studies.[ 13, 14 ]. p<0.001) were independently associated with delayed hospital discharge (Fig 3; S5 Table). HRR was dose-dependently associated with length of hospital stay (S6 Table). Sensitivity analysis Patients with HRR 12 beats.min-1 remained more likely to sustain morbidity within 5 days of surgery (OR:1.38 (1.12?1.69); p<0.001; S1 Fig), independent of the presence/absence of negatively prognostic CPET factors VO2 peak 14 ml kg min and/or VE/VCO2 ratio 34.[ 28 ] A similar relationship was observed for patients receiving beta-blockers (OR:1.24 (0.83?1.86). Mortality was 0.7% in patients with preoperative HRR <12, compared to 0.3% in those with HRR>12 (odds ratio:2.7 (95%CI:0.7?10.7)). Discussion The principal finding of these two large generalisable studies is that parasympathetic dysfunction?as quantified by delayed HRR after preoperative exercise- is more frequently associated with morbidity within 5 days of surgery. The broad range, and temporal pattern, of morbidity after surgery is consistent with laboratory/translational work highlighting that loss of vagal 8 / 15 Relative risk (95%CI) activity promotes multiple organ dysfunction. These data reinforce prognostic studies from the general population, where delayed HRR 12 beats.minute-1 is a strong independent predictor of all-cause mortality in apparently otherwise healthy middle-aged subjects.[ 13, 14 ] Fig 3. Delayed heart rate and postoperative outcome. Unadjusted Kaplan-Meier estimator for all surgery types showed that delayed heart rate recovery was associated with prolonged hospital stay (HR: 1.15 (95%CI:1.05?1.26); p = 0.0008). 9 / 15 Taken together, these data suggest that cardiac vagal dysfunction represents a distinct endo type for morbidity after major surgery. Vagal reactivation is the primary mechanism underlying deceleration of heart rate after exercise, as demonstrated by its blockade in humans by atropine.[ 17 ] Athletes exhibit accelerated, vagally mediated heart rate recovery after exercise, in contrast to the blunted response observed in patients with heart failure.[ 17 ] Heart rate control is attributable to neuronal substrate within the nucleus ambiguous. The other main parasympathetic brainstem substrate, the dorsal vagal motor nucleus, also regulates cardiac ventricular function, as well as innervating multiple other organs.[ 4 ] The development of morbidity across multiple organs suggests that a broad deficiency in parasympathetic neurons may be present in many patients. However, we cannot rule out that many of the morbidities over-represented in patients with delayed heart rate recovery are linked indirectly through non-neural mechanisms. For example, hypotension is more common in patients with delayed heart rate recovery, which in turn may predispose to acute renal dysfunction.[ 29 ] Nevertheless, parasympathetic neural activity contributes to a diverse range of physiological functions, with a broader role being plausible in pathological states. Laboratory data shows that vagal neuromodulation of cardiomyocyte, gastrointestinal, renal, lung, neurological and immune cells limits tissue injury in a range of pathological settings. [ 7, 30, 31 ] Lack of cardiac vagal activity promotes arrhythmias,[ 32 ] perhaps in part due to the detrimental effects of ongoing systemic inflammation.[ 33 ] Further reductions in parasympathetic activity induced by anaesthetic agents,[ 34 ] opioid analgesics[ 35 ] and inflammatory modulation of afferent autonomic inputs[ 36 ] following surgery may therefore contribute to postoperative morbidity.[ 2 ] Our data support the findings of a small study in patients undergoing thoracic surgery for lung cancer, where a threshold value of HRR 12-beat.min-1 following the six-minute walk test was associated with postoperative cardiopulmonary complications.[ 37 ] In cardiac failure, loss of parasympathetic activity is a consistent predictor of accelerated mortality.[ 38 ] Failure to improve autonomic function- despite optimal medical therapy- is associated with higher mortality.[ 39 ] Patients presenting for major surgery, who are frequently deconditioned as a result of cancer and/or systemic inflammation,[ 40 ] appear to share similar strikingly similar pathophysiological features.[ 28 ] Impaired heart rate recovery is independent of clinically-defined cardiac comorbidity and is not influenced by drug therapy including continuation of ?-blockade.[ 12, 40 ] A strength of this study is the blinding of HRR data to participants and clinicians. The prospective, international multi-centre study design suggests that these data are generalisable. Our findings are limited by the morbidity measure used, given the sensitivity of the POMS for many morbidities after major surgery. This limitation was partly addressed by tracking serial morbidity, which allowed us to assess time to become morbidity free. This measure is more mechanistically informative, as it reflects morbidity over time- as opposed to a snapshot primary outcome which may be misleading. The systematic grading of severity was captured partially for some POMS-domains, and is likely to have provided a richer, more granular understanding. The use of organ-specific biomarkers, as we have found with high-sensitivity troponin,[ 12 ] may reveal further mechanistic insights. The choice of HRR threshold, and timepoint after cessation of exercise, requires further study. However, we used a standardised exercise protocol, combined with the most stringent and widely adopted threshold of HRR described thus far.[ 41 ] Moreover, in mechanistic work, we found a strong association between the HRR threshold used and a pathological cellular phenotype.[ 42 ] In summary, delayed HRR is independently associated with excess postoperative morbidity leading to prolonged admission. Preoperative parasympathetic dysfunction represents a distinct endotype for morbidity after major surgery. 10 / 15 Supporting information S1 File. Minimum dataset: Primary outcomes. (XLSX) S1 Table. POMS-defined morbidity. (DOCX) S2 Table. POMS-defined morbidity on postoperative days 3 and 5. (DOCX) S3 Table. Factors associated with patients being free of morbidity within 5 days of surgery. (DOCX) S4 Table. Relationship between magnitude of HRR and factors associated with patients being free of morbidity within 5 days of surgery. (DOCX) S5 Table. Factors associated with delayed discharge from hospital after surgery. (DOCX) S6 Table. Relationship between magnitude of HRR and other factors associated with more rapid discharge from hospital after surgery. (DOCX) S1 Fig. Serial changes in POMS-defined morbidity on postoperative days 3 and 5. (DOCX) S2 Fig. Delayed heart rate recovery and CPET markers of severe cardiac failure. (DOCX) S3 Fig. Delayed heart rate and postoperative outcome. (DOCX) Acknowledgments METS Study Investigators 11 / 15 POM-HR study investigators Derriford Hospital, Plymouth Hospitals NHS Trust & Peninsula Schools of Medicine and Dentistry, Plymouth University, United Kingdom: Gary Minto, Angela King, Claire Pollak, Claire Williams, Abigail Patrick, Claire West. Royal Bournemouth Hospital, United Kingdom: Emma Vickers, Richard Green, Martin Clark. University College London Hospitals NHS Trust, London, United Kingdom: Gareth L Ackland, Robert CM Stephens, Amy Jones, James Otto, Anna Lach, Ana Gutierrez del Arroyo, John Whittle, Laura Gallego Paredes, Anna Reyes, Andrew Toner. Royal Preston Hospital, Lancashire Teaching Hospitals NHS Foundation Trust, United Kingdom: Alexandra Williams, Thomas Owen. 12 / 15 Royal Surrey County Hospital, Guildford, United Kingdom: Pradeep Pradhu, Daniel Hull, Laura Montague. Author Contributions Conceptualization: Gareth L. Ackland. Data curation: Tom E. F. Abbott, Martin Clark, Thomas Owen, Pradeep Prabhu, Joseph A. Reynolds, Brian H. Cuthbertson, Duminda Wijesundera, Rupert M. Pearse. Formal analysis: Gareth L. Ackland, Tom E. F. Abbott, Shaun M. May, Brian H. Cuthbertson, Duminda Wijesundera, Rupert M. Pearse. Wijesundera, Rupert M. Pearse. Funding acquisition: Gareth L. Ackland, Tom E. F. Abbott, Brian H. Cuthbertson, Duminda Investigation: Tom E. F. Abbott, Gary Minto, Martin Clark, Thomas Owen, Pradeep Prabhu, Brian H. Cuthbertson, Duminda Wijesundera, Rupert M. Pearse. Methodology: Gareth L. Ackland, Tom E. F. Abbott, Brian H. Cuthbertson, Duminda Wijesundera, Rupert M. Pearse. Wijesundera, Rupert M. Pearse. Project administration: Tom E. F. Abbott, Gary Minto, Brian H. Cuthbertson, Duminda Resources: Gareth L. Ackland, Brian H. Cuthbertson, Duminda Wijesundera, Rupert M. Pearse. Pearse. Pearse. Software: Brian H. Cuthbertson, Duminda Wijesundera, Rupert M. Pearse. Supervision: Gareth L. Ackland, Martin Clark, Thomas Owen, Pradeep Prabhu, Brian H. Cuthbertson, Duminda Wijesundera, Rupert M. Pearse. Validation: Joseph A. Reynolds, Brian H. Cuthbertson, Duminda Wijesundera, Rupert M. Visualization: Tom E. F. Abbott, Joseph A. Reynolds. Writing ? original draft: Gareth L. Ackland, Tom E. F. Abbott, Brian H. Cuthbertson. Writing ? review & editing: Gareth L. Ackland, Gary Minto, Shaun M. May, Rupert M. 13 / 15 14 / 15 1. Amar D , Fleisher M , Pantuck CB , Shamoon H , Zhang H , Roistacher N , et al. Persistent alterations of the autonomic nervous system after noncardiac surgery . Anesthesiology . 1998 ; 89 ( 1 ): 30 - 42 . Epub 1998/07/17. https://doi.org/10.1097/ 00000542 -199807000-00008 PMID: 9667291 . 2. Karmali S , Jenkins N , Sciuso A , John J , Haddad F , Ackland GL . Randomized controlled trial of vagal modulation by sham feeding in elective non-gastrointestinal (orthopaedic) surgery . Br J Anaesth . 2015 ; 115 : 727 - 35 . https://doi.org/10.1093/bja/aev283 PMID: 26323293 . 3. Johnston GR , Webster NR . Cytokines and the immunomodulatory function of the vagus nerve . Br J Anaesth . 2009 ; 102 ( 4 ): 453 - 62 . Epub 2009/03/05. https://doi.org/10.1093/bja/aep037 aep037 [pii]. PMID: 19258380. 4. Ackland GL , Gourine AV . Cardiac vagus and exercise . Physiology (Bethesda) . 2019 ; 34 ( 1 ): 71 - 80 . Epub https://doi.org/10.1152/physiol.00041.2018 PubMed Central PMCID : PMC30540229 . PMID: 30540229 5. Chavan SS , Pavlov VA , Tracey KJ . Mechanisms and Therapeutic Relevance of Neuro-immune Communication . Immunity . 2017 ; 46 ( 6 ): 927 - 42 . https://doi.org/10.1016/j.immuni. 2017 . 06 .008 PMID: 28636960; PubMed Central PMCID : PMC5578398 . 6. Mastitskaya S , Marina N , Gourine A , Gilbey MP , Spyer KM , Teschemacher AG , et al. Cardioprotection evoked by remote ischaemic preconditioning is critically dependent on the activity of vagal pre-ganglionic neurones . Cardiovasc Res . 2012 ; 95 ( 4 ): 487 - 94 . https://doi.org/10.1093/cvr/cvs212 PMID: 22739118; PubMed Central PMCID : PMC3422080 . 7. dos Santos CC , Shan Y , Akram A , Slutsky AS , Haitsma JJ . Neuroimmune regulation of ventilatorinduced lung injury . American journal of respiratory and critical care medicine . 2011 ; 183 ( 4 ): 471 - 82 . Epub 2010/09/28. https://doi.org/10.1164/rccm.201002-0314OC 201002 - 0314OC [pii]. PMID: 20870758. 8. Yeboah MM , Xue X , Duan B , Ochani M , Tracey KJ , Susin M , et al. Cholinergic agonists attenuate renal ischemia-reperfusion injury in rats . Kidney Int . 2008 ; 74 ( 1 ): 62 - 9 . https://doi.org/10.1038/ki. 2008 .94 PMID: 18401335 9. Inoue T , Abe C , Sung SJ , Moscalu S , Jankowski J , Huang L , et al. Vagus nerve stimulation mediates protection from kidney ischemia-reperfusion injury through alpha7nAChR+ splenocytes . J Clin Invest . 2016 ; 126 : 1939 - 52 . https://doi.org/10.1172/JCI83658 PMID: 27088805 . 10. Kirchner A , Birklein F , Stefan H , Handwerker HO . Left vagus nerve stimulation suppresses experimentally induced pain . Neurology . 2000 ; 55 ( 8 ): 1167 - 71 . https://doi.org/10.1212/wnl.55.8.1167 PMID: 11071495 . 11. Karmali S , Jenkins N , Sciusco A , John J , Haddad F , Ackland GL , et al. Randomized controlled trial of vagal modulation by sham feeding in elective non-gastrointestinal (orthopaedic) surgery . Br J Anaesth . 2015 ; 115 ( 5 ): 727 - 35 . https://doi.org/10.1093/bja/aev283 PMID: 26323293 . 12. Abbott TEF , Pearse RM , Cuthbertson BH , Wijeysundera D , Ackland GL . Cardiac vagal dysfunction and myocardial injury after non-cardiac surgery: a planned secondary analysis of the METS study . Br J Anaesth . 2018 ;BJA-2018 - 01229 -MA081; accepted for publication. 13. Cole CR , Blackstone EH , Pashkow FJ , Snader CE , Lauer MS . Heart-rate recovery immediately after exercise as a predictor of mortality . The New England journal of medicine . 1999 ; 341 ( 18 ): 1351 - 7 . Epub 1999/10/28. https://doi.org/10.1056/NEJM199910283411804 PMID: 10536127 . 14. Jouven X , Empana JP , Schwartz PJ , Desnos M , Courbon D , Ducimetiere P . Heart-rate profile during exercise as a predictor of sudden death . N Engl J Med . 2005 ; 352 ( 19 ): 1951 - 8 . Epub 2005/05/13. 352/ 19/1951 [pii] https://doi.org/10.1056/NEJMoa043012 PMID: 15888695 . 15. Pierpont GL , Voth EJ . Assessing autonomic function by analysis of heart rate recovery from exercise in healthy subjects . Am J Cardiol . 2004 ; 94 ( 1 ): 64 - 8 . https://doi.org/10.1016/j.amjcard. 2004 . 03 .032 PMID: 15219511 . 16. Cahalin LP , Forman DE , Chase P , Guazzi M , Myers J , Bensimhon D , et al. The prognostic significance of heart rate recovery is not dependent upon maximal effort in patients with heart failure . Int J Cardiol . 2013 ; 168 ( 2 ): 1496 - 501 . https://doi.org/10.1016/j.ijcard. 2012 . 12 .102 PMID: 23391698 . 17. Imai K , Sato H , Hori M , Kusuoka H , Ozaki H , Yokoyama H , et al. Vagally mediated heart rate recovery after exercise is accelerated in athletes but blunted in patients with chronic heart failure . J Am Coll Cardiol . 1994 ; 24 ( 6 ): 1529 - 35 . Epub 1994/11/15. https://doi.org/10.1016/ 0735 - 1097 ( 94 ) 90150 - 3 PMID: 7930286 . 18. Gourine AV , Machhada A , Trapp S , Spyer KM . Cardiac vagal preganglionic neurones: An update . Auton Neurosci . 2016 ; 199 : 24 - 8 . https://doi.org/10.1016/j.autneu. 2016 . 06 .003 PMID: 27396874 . 2016 ; 6(3):e010359 . https://doi.org/10.1136/bmjopen-2015-010359 PMID: 26969643; PubMed Central PMCID : PMC4800144 . Wijeysundera DN , Pearse RM , Shulman MA , Abbott TEF , Torres E , Ambosta A , et al. Assessment of functional capacity before major non-cardiac surgery: an international, prospective cohort study . Lancet . 2018 ; 391 ( 10140 ): 2631 - 40 . https://doi.org/10.1016/S0140- 6736 ( 18 ) 31131 - 0 PMID: 30070222 . 21. Levett DZH , Jack S , Swart M , Carlisle J , Wilson J , Snowden C , et al. Perioperative cardiopulmonary exercise testing (CPET): consensus clinical guidelines on indications, organization, conduct, and physiological interpretation . Br J Anaesth . 2018 ; 120 ( 3 ): 484 - 500 . https://doi.org/10.1016/j.bja. 2017 . 10 .020 PMID: 29452805 . 22. Cole CR , Blackstone EH , Pashkow FJ , Snader CE , Lauer MS . Heart-rate recovery immediately after exercise as a predictor of mortality . The New England journal of medicine . 1999 ; 341 ( 18 ): 1351 - 7 . https://doi.org/10.1056/NEJM199910283411804 PMID: 10536127 . 23. Vivekananthan DP , Blackstone EH , Pothier CE , Lauer MS . Heart rate recovery after exercise is a predictor of mortality, independent of the angiographic severity of coronary disease . J Am Coll Cardiol . 2003 ; 42 ( 5 ): 831 - 8 . https://doi.org/10.1016/s0735- 1097 ( 03 ) 00833 - 7 PMID: 12957428 . 24. Abbott TEF , Minto G , Lee AM , Pearse RM , Ackland G . Elevated preoperative heart rate is associated with cardiopulmonary and autonomic impairment in high-risk surgical patients . British journal of anaesthesia . 2017 ; 119 ( 1 ): 87 - 94 . https://doi.org/10.1093/bja/aex164 PMID: 28974075 25. Karnik RS , Lewis W , Miles P , Baker L . The effect of beta-blockade on heart rate recovery following exercise stress echocardiography . Prev Cardiol . 2008 ; 11 ( 1 ): 26 - 8 . PMID: 18174788 . 26. West MA , Lythgoe D , Barben CP , Noble L , Kemp GJ , Jack S , et al. Cardiopulmonary exercise variables are associated with postoperative morbidity after major colonic surgery: a prospective blinded observational study . Br J Anaesth . 2014 ; 112 ( 4 ): 665 - 71 . https://doi.org/10.1093/bja/aet408 PMID: 24322573 . 27. Ackland GL , Moran N , Cone S , Grocott MP , Mythen MG . Chronic kidney disease and postoperative morbidity after elective orthopedic surgery . Anesth Analg . 2011 ; 112 ( 6 ): 1375 - 81 . https://doi.org/10. 1213/ANE.0b013e3181ee8456 PMID: 20807976 . 28. Abbott TEF , Minto G , Lee AM , Pearse RM , Ackland GL , Pom-Hr P-O , et al. Elevated preoperative heart rate is associated with cardiopulmonary and autonomic impairment in high-risk surgical patients . Br J Anaesth . 2017 ; 119 ( 1 ): 87 - 94 . https://doi.org/10.1093/bja/aex164 PMID: 28974075 . 29. Sun LY , Wijeysundera DN , Tait GA , Beattie WS . Association of intraoperative hypotension with acute kidney injury after elective noncardiac surgery . Anesthesiology . 2015 ; 123 ( 3 ): 515 - 23 . https://doi.org/ 10.1097/ALN.0000000000000765 PMID: 26181335 . 30. Cheyuo C , Jacob A , Wu R , Zhou M , Coppa GF , Wang P. The parasympathetic nervous system in the quest for stroke therapeutics . J Cereb Blood Flow Metab . 2011 ; 31 ( 5 ): 1187 - 95 . Epub 2011/03/03. https://doi.org/10.1038/jcbfm. 2011 . 24 jcbfm201124 [pii]. PMID: 21364605; PubMed Central PMCID : PMC3099641 . 31. Kox M , Pompe JC , Peters E , Vaneker M , van der Laak JW , van der Hoeven JG , et al. alpha7 nicotinic acetylcholine receptor agonist GTS-21 attenuates ventilator-induced tumour necrosis factor-alpha production and lung injury . Br J Anaesth . 2011 ; 107 ( 4 ): 559 - 66 . Epub 2011/07/21. https://doi.org/10.1093/ bja/aer202 aer202 [pii]. PMID: 21771746. 32. Schwartz PJ , La Rovere MT , De Ferrari GM , Mann DL . Autonomic modulation for the management of patients with chronic heart failure . Circ Heart Fail . 2015 ; 8 ( 3 ): 619 - 28 . https://doi.org/10.1161/ CIRCHEARTFAILURE.114. 001964 PMID: 25991804 . 33. Andersson U , Tracey KJ . Neural reflexes in inflammation and immunity . J Exp Med . 2012 ; 209 ( 6 ): 1057 - 68 . Epub 2012/06/06. https://doi.org/10.1084/jem.20120571 jem.20120571 [pii]. PMID: 22665702; PubMed Central PMCID : PMC3371736 . 34. Wang X , Huang ZG , Gold A , Bouairi E , Evans C , Andresen MC , et al. Propofol modulates gamma-aminobutyric acid-mediated inhibitory neurotransmission to cardiac vagal neurons in the nucleus ambiguus . Anesthesiology . 2004 ; 100 ( 5 ): 1198 - 205 . https://doi.org/10.1097/ 00000542 -200405000-00023 PMID: 15114218 . 35. Browning KN , Kalyuzhny AE , Travagli RA . Opioid peptides inhibit excitatory but not inhibitory synaptic transmission in the rat dorsal motor nucleus of the vagus . J Neurosci . 2002 ; 22 ( 8 ): 2998 - 3004 . doi: 20026224 PMID: 11943802; PubMed Central PMCID : PMC3062482 . 36. Ackland GL , Kazymov V , Marina N , Singer M , Gourine AV . Peripheral neural detection of danger-associated and pathogen-associated molecular patterns . Crit Care Med . 2013 ; 41 ( 6 ): e85 - 92 . Epub 2013/ 03/20. https://doi.org/10.1097/CCM.0b013e31827c0b05 PMID: 23507714 . 37. Ha D , Choi H , Zell K , Raymond DP , Stephans K , Wang XF , et al. Association of impaired heart rate recovery with cardiopulmonary complications after lung cancer resection surgery . J Thorac Cardiovasc Surg . 2015 ; 149 ( 4 ): 1168 - 73 e3. https://doi.org/10.1016/j.jtcvs. 2014 . 11 .037 PMID: 25583106 . 38. Schwartz PJ , De Ferrari GM. Sympathetic-parasympathetic interaction in health and disease: abnormalities and relevance in heart failure . Heart Fail Rev . 2011 ; 16 ( 2 ): 101 - 7 . Epub 2010/06/26. https://doi. org/10.1007/s10741-010-9179-1 PMID: 20577900 . 39. Nolan J , Batin PD , Andrews R , Lindsay SJ , Brooksby P , Mullen M , et al. Prospective study of heart rate variability and mortality in chronic heart failure: results of the United Kingdom heart failure evaluation and assessment of risk trial (UK-heart) . Circulation . 1998 ; 98 ( 15 ): 1510 - 6 . https://doi.org/10.1161/01. cir. 98 .15.1510 PMID: 9769304 40. Ackland GL , Minto G , Clark M , Whittle J , Stephens RCM , Owen T , et al. Autonomic regulation of systemic inflammation in humans: A multi-center, blinded observational cohort study . Brain Behav Immun . 2018 ; 67 : 47 - 53 . https://doi.org/10.1016/j.bbi. 2017 . 08 .010 PMID: 28807718 . 41. Adabag S , Pierpont GL . Exercise heart rate recovery: analysis of methods and call for standards . Heart . 2013 ; 99 ( 23 ): 1711 - 2 . https://doi.org/10.1136/heartjnl-2013 -303863 PMID: 23591670. 42. Ackland GL , Whittle J , Toner A , Machhada A , Del Arroyo AG , Sciuso A , et al. Molecular Mechanisms Linking Autonomic Dysfunction and Impaired Cardiac Contractility in Critical Illness . Crit Care Med . 2016 ; March 4 . https://doi.org/10.1097/CCM.0000000000001606 PMID: 26950003 .


This is a preview of a remote PDF: https://journals.plos.org/plosone/article/file?id=10.1371/journal.pone.0221277&type=printable

Gareth L. Ackland, Tom E. F. Abbott, Gary Minto, Martin Clark, Thomas Owen, Pradeep Prabhu, Shaun M. May, Joseph A. Reynolds, Brian H. Cuthbertson, Duminda Wijesundera, Rupert M. Pearse, for the METS and POM-HR Study Investigators. Heart rate recovery and morbidity after noncardiac surgery: Planned secondary analysis of two prospective, multi-centre, blinded observational studies, PLOS ONE, 2019, DOI: 10.1371/journal.pone.0221277