Share this article on:

The Effect of Hypothyroidism on a Composite of Mortality, Cardiovascular and Wound Complications After Noncardiac Surgery: A Retrospective Cohort Analysis

Komatsu, Ryu MD*; You, Jing MS†‡; Mascha, Edward J. PhD†‡; Sessler, Daniel I. MD; Kasuya, Yusuke MD; Turan, Alparslan MD

doi: 10.1213/ANE.0000000000000805
Patient Safety: Research Report

BACKGROUND: We tested the hypothesis that hypothyroidism, as defined by thyroid-stimulating hormone (TSH) concentration, is associated with a severity-weighted composite of mortality and major cardiovascular and infectious complications after noncardiac surgery.

METHODS: In this retrospective cohort study, we evaluated adults at the Cleveland Clinic Main Campus between 2005 and 2012, who had had available TSH concentrations within the 6 months before noncardiac surgery. Patients were categorized as (1) hypothyroid (patients who had diagnosis of hypothyroidism any time prior to surgery and increased TSH value (> 5.5 mIU/L) within 6 months prior to surgery); (2) treated (hypothyroid diagnosis and normal TSH concentrations [0.4–5.5 mIU/L]); and (3) euthyroid (no hypothyroid diagnosis and normal TSH concentrations). We conducted pairwise comparisons among the 3 groups using inverse propensity score weighting to control for observed confounding variables. Average relative effect generalized estimating equation model was used for the primary outcome composite of in-hospital cardiovascular morbidity, surgical wound complication or infection, and mortality. Logistic regression and Cox proportional hazards regression were used for secondary outcomes of intraoperative vasopressor use and duration of hospitalization, respectively.

RESULTS: We identified 800 hypothyroid patients (median TSH: 8.6 mIU/L [Q1, Q3: 6.5, 13.0]), 1805 treated patients (2.0 mIU/L [1.1, 3.2]), and 5612 euthyroid patients (1.7 mIU/L [1.1, 2.6]). There were no significant differences among the hypothyroid, treated, and euthyroid patients on the primary composite outcome (all P values ≥0.30). Hypothyroid patients were slightly more likely to receive vasopressor during surgery than either treated (odds ratio, 1.17; 99.2% confidence interval [CI], 1.01–1.36) or euthyroid (odds ratio, 1.12; 99.2% CI, 1.02–1.24) patients. Furthermore, hypothyroid patients were slightly but significantly less likely to be discharged at any given postoperative time than treated patients (hazard ratio, 0.92; 99.2% CI, 0.86–0.99).

CONCLUSIONS: Hypothyroidism was not associated with worse postoperative mortality, wound, or cardiovascular outcomes in noncardiac patients. Thus, postponing surgery to initiate thyroid replacement therapy in patients with hypothyroidism seems unnecessary.

From the *Anesthesiology Institute, Cleveland Clinic, Cleveland, Ohio; Departments of Quantitative Health Sciences and Outcomes Research, Cleveland Clinic, Cleveland, Ohio.

Yusuke Kasuya, MD, is currently affiliated with the Department of Anesthesiology, Tokyo Women’s Medical University Hospital, Tokyo, Japan.

Accepted for publication February 17, 2015.

Funding: This research was supported solely by internal sources.

The authors declare no conflicts of interest.

This report was previously presented, in part, at the American Society of Anesthesiologists Annual Meeting, San Francisco, CA, October 2013.

Address correspondence to Ryu Komatsu, MD, Anesthesiology Institute, Cleveland Clinic, 9500 Euclid Ave., E30, Cleveland, OH 44195. Address e-mail to ryukomatsu80@gmail.com.

Hypothyroidism is common, with 2% to 10% of the population having increased serum thyroid-stimulating hormone (TSH) concentrations.1–6 Remarkably, approximately 20% of patients taking thyroid hormone replacement remain hypothyroid.6,7 Chronic hypothyroidism causes bradycardia, increased systemic vascular resistance, and decreased cardiac output. Both systolic and diastolic myocardial functions are impaired, and it occasionally causes congestive heart failure even in patients without structural heart disease.8 A prominent electrocardiographic feature is a prolonged QT interval that promotes ventricular tachycardia, particularly torsades de pointe.9 Even subclinical hypothyroidism may have adverse effects on the cardiovascular system.10,11 Thyroid hormones also play a part in the tissue healing process, moderating fibroblast function;12–14 consequently, hypothyroidism is associated with postoperative cutaneous and visceral wound complications.15–19

Several small studies have evaluated perioperative outcomes in hypothyroid patients with largely contradictory results. For example, Myerowitz et al.20 reported a lower cardiac index in patients with low serum thyroxin concentrations in comparison with those with high-normal concentrations. Ladenson et al.21 reported increased heart failure, more gastrointestinal complications, and delayed recovery from anesthesia in hypothyroid patients. Park et al.22 found that the incidence of postoperative atrial fibrillation was increased in patients with subclinical hypothyroidism. Finally, Ladenson et al.21 observed that hypothyroid patients were more likely to experience intraoperative hypotension and postoperative neuropsychiatric complications. In contrast, other studies did not identify increased postoperative adverse events.23–26

In the absence of relevant large-scale randomized controlled trials involving hypothyroid patients, it remains unclear whether hypothyroidism is associated with worse perioperative outcomes. Therefore, we evaluated the relationship between hypothyroidism and presumably major complications in noncardiac surgical patients, using a retrospective cohort study design. Specifically, we tested the primary hypothesis that patients with laboratory-documented hypothyroidism, as indicated by preoperative TSH concentrations >5.5 mIU/L, were at a greater risk for a composite of mortality and Agency for Healthcare Research & Quality–defined serious cardiovascular and surgical wound complications. The reference groups included patients with normal TSH concentrations, whether occurring naturally or corrected by thyroid hormone supplementation. Our secondary hypotheses were that patients with laboratory-documented hypothyroidism experience more frequent intraoperative hypotension requiring treatment with vasopressors and remain hospitalized longer than those who are euthyroid.

Back to Top | Article Outline

METHODS

With approval from the Cleveland Clinic IRB and waiver of informed consent, we evaluated electronic records in the Perioperative Health Documentation System Registry at the Cleveland Clinic for adults who had noncardiac surgery at the Cleveland Clinic Main Campus between January 11, 2005, and August 24, 2012. We excluded patients with ASA physical status exceeding IV. We also excluded patients without a recorded TSH concentration within the 6 months before surgery or with a TSH measurement <0.4 mIU/L. TSH concentrations were measured using immune assay (cobas®, Roche, Indianapolis, IN).

Qualifying patients were divided into 3 groups: (1) a preoperative diagnosis of hypothyroidism and TSH concentration >5.5 mIU/L, irrespective of thyroid hormone supplementation (hypothyroid); (2) a recorded preoperative diagnosis of hypothyroidism and a normal preoperative TSH concentration (i.e., 0.4 ≤ TSH ≤ 5.5 mIU/L), irrespective of thyroid hormone supplementation (treated); and (3) no preoperative diagnosis of hypothyroidism, not receiving thyroid hormone supplementation, and normal preoperative TSH concentration (euthyroid). If multiple measurements were obtained within 6 months before surgery, the value measured closest to the date of surgery was used. We used TSH concentrations to distinguish thyroid status because few had available T3 or T4 test results.

Back to Top | Article Outline

Propensity Score Methods to Control for Observed Confounding

We conducted pairwise comparisons among the 3 hypothyroid conditions (i.e., hypothyroid, treated, and euthyroid) on outcomes (detailed below). We used the inverse propensity score weighting method to control for observed confounding.27 For each comparison, this involved first fitting a multivariable logistic regression model in which the outcome variable was the most severe group being compared (e.g., the outcome variable was the hypothyroid group when comparing hypothyroid with euthyroid), and the independent variables were all available potentially confounding variables (Table 1). From this model, propensity scores (i.e., the probability of being in the most severe group being compared) were estimated for each patient. Pairwise comparisons among hypothyroid, treated, and euthyroid on outcome were made in separate logistic regression models in which each observation was weighted by the inverse of the propensity score pertaining to the most severe group being compared.

Table 1

Table 1

Success of the control for confounding was assessed by comparing groups on all the potential confounding variables used to construct the propensity scores with the standardized difference after weighting observations by the inverse of the relevant propensity score. Variables with an absolute standardized difference >0.083 (i.e.,

CV

CV

) between hypothyroid and treated groups, >0.074 (i.e.,

CV

CV

) between hypothyroid and euthyroid groups, and >0.053 (i.e.,

CV

CV

) between treated and euthyroid groups were considered as imbalanced28 and adjusted for in the relevant analyses.

We conducted a sensitivity analysis using the multiple propensity score method29 to adjust for confounding across the 3 groups of interest simultaneously. First, we fitted a multiple-response (polytomous) logistic regression model in which the outcome variable was the study group and the independent variables were all potentially confounding variables listed in Table 1. A predicted set of propensity scores (1 for each group) was estimated from this model for each patient and adjusted for in the models, assessing the difference among hypothyroid, treated, and euthyroid on primary outcomes. This method is weaker than the primary analysis presented earlier, because the adjustment for confounding by simply adjusting for the propensity scores is not optimal. Also, it does not allow visualization of the achieved balance among groups on confounding variables.

We also conducted a second sensitivity analysis using a propensity score matching method to adjust for the potential confounders. First, a 1-to-2 propensity score matching hypothyroid and treated patients was obtained as follows: we estimated the probability (i.e., the propensity score) of being hypothyroid (versus treated) using logistic regression based on the potentially confounding variables listed in Table 1. A 1-to-2 greedy distance matching algorithm (using a maximum allowable logit of propensity score within 0.2 SDs28) was used. Similarly, we obtained 1-to-4 matched hypothyroid and euthyroid patients and 1-to-2 matched treated and euthryoid patients. Then, the comparisons were used for comparing the matched groups. The drawback of this method is that different subsets of patients were used for different comparisons.

Back to Top | Article Outline

Primary Outcome

Our primary outcome was a composite of major in-hospital cardiovascular morbidity, surgical wound complication or infection morbidity, and mortality (as defined in Table 2). Each of the 3 outcome components for a patient had a value 1 or 0 depending on whether the patient experienced any of the specific events comprising that component, i.e., a component-specific “composite.” However, the outcome components would not likely be considered by researchers or patients to have exactly the same severity, which cannot be accounted for in the “any-versus-none” approach. Therefore, we used a multivariate (i.e., 1 record/outcome/patient) analysis to simultaneously capture information on individual outcome components for a patient, the correlations among components, and allow severity weighting for the outcome components.

Table 2

Table 2

By using a questionnaire, the clinical severity weight was obtained from independent survey results of clinicians’ perceived severity of each of the 3 components of morbidity from 11 attending anesthesiologists at our institution who were otherwise not involved in the study. Perceived clinical severity of each component was expressed on a 0 to 100 score (0, negligible; 100, most severe). Weights were determined as the average of ratings among 11 staff anesthesiologists in our prestudy survey. The average ratings were 100 (SD, 0) for mortality, 65 (SD, 13; 95% confidence interval [CI], 63–67) for cardiovascular morbidity, and 47 (SD, 16; 95% CI, 45–49) for surgical wound complication or infection. The ratings for cardiovascular morbidity and surgical wound complication or infection were strongly consistent with each other, indicated by a standardized Cronbach α coefficient of 0.91.

We conducted pairwise comparisons among the 3 hypothyroid conditions on the set of primary outcome components. For each comparison, we estimated the average relative effect across the 3 individual components (severity-weighted average of the 3 log-odds ratios [ORs]), using a generalized estimating equation multivariate model with an unstructured covariance matrix.30 The analyses were weighted by the inverse of the propensity score pertaining to the most severe group being compared and adjusted for any imbalanced covariables between the compared groups with control for observed confounding.

In contrast to the more common collapsed composite method that compares groups on whether any (versus none) of the components were observed for a patient, the average relative effect generalized estimating equation method that we used captures complete information on each component for each patient, adjusts for the correlation among components, and is not driven by component(s) with the highest frequency.31 A related distinct-effects generalized estimating equation model was used to assess the heterogeneity of the hypothyroid effect across the components of the primary outcome by testing the hypothyroid-by-outcome interaction.31

Back to Top | Article Outline

Secondary Outcomes

Hypothyroid, treated, and euthyroid patients were compared on intraoperative vasopressor use (i.e., dobutamine, dopamine, ephedrine, epinephrine, norepinephrine, phenylephrine, or vasopressin) using multivariable logistic regression and on duration of hospitalization using Cox proportional hazards regression on time to discharge alive, each weighting by the inverse of the propensity score pertaining to the most severe group being compared and adjusting for any imbalanced covariables. Because some patients died in hospital, analyzing time to discharge in the duration of hospitalization analysis would create bias, because early deaths would look favorable. Instead, we analyzed time to discharge alive and considered the in-hospital deaths as nonevents, censoring them at longest observed alive-discharge length of stay.

Intraoperative hemodynamic monitoring data were acquired from our electronic anesthesia record-keeping system off of the anesthesia monitor. Artifact values including extreme values of arterial blood pressures, heart rates, electrocardiogram, and pulse oximeter oxygen saturation values caused by electrocautery, movements, and transducer failure for invasive arterial, central venous, and pulmonary artery pressure values were removed from the Perioperative Health Documentation System Registry by a contracted device company (Aspect Medical, Norwood, MA, now part of Covidien, Minneapolis, MN). Arterial blood pressure in patients with arterial catheters was recorded every minute and at 1- to 5-minute intervals in others. Average systolic and diastolic pressures were computed during each period (i.e., start of case to induction, induction to tracheal intubation, intubation to incision, incision to closing, closing to emergence, and emergence to end of case) in each patient; minimum systolic and diastolic blood pressures during each period in each patient were averaged and reported.

Bonferroni correction was used to control the type I error at 0.05, so that the significance criterion for each of 3 pairwise comparisons on the primary outcome was P < 0.017 (i.e., 0.05/3) and for each pairwise comparison on 2 secondary analyses was P < 0.008 (i.e., 0.05/6, 3 pairwise comparisons × 2 outcomes). Thus, 98.3% and 99.2% CIs were reported, respectively. SAS software version 9.3 (SAS Institute, Cary, NC) and R software version 2.12.1 (The R Foundation for Statistical Computing, Vienna, Austria) were used for all statistical analysis.

Back to Top | Article Outline

Sample Size and Power

Eight hundred hypothyroid, 1805 treated, and 5612 euthyroid patients in our registry fulfilled the study inclusion criteria. We a priori expected an incidence of approximately 9% for each of the wound complication/infection and cardiovascular outcomes for the euthyroid group. With these sample sizes and expected control group incidences, we had approximately 90% power at the overall 0.05 significance level to detect ORs of ≥1.4 between the hypothyroid group and each of the treated and euthyroid groups.

Back to Top | Article Outline

RESULTS

There were 134,607 adults who had noncardiac surgery at the Cleveland Clinic Main Campus between 2005 and 2012. Among 8217 qualifying adults who had TSH concentrations recorded within 6 months of surgery, there were 800 hypothyroid patients, 1805 treated patients, and 5612 euthyroid patients (Fig. 1). The corresponding median TSH concentrations (within 6 months of the date of surgery) were 8.6 mIU/L (interquartile range, 6.5–13.0 mIU/L) for hypothyroid patients, 2.0 mIU/L (1.1–3.2 mIU/L) for treated patients, and 1.7 mIU/L (1.1–2.6 mIU/L) for euthyroid patients. One hundred eighty-nine (24%) of the hypothyroid patients and 485 (27%) of the treated patients were taking thyroid hormone supplementation.

Figure 1

Figure 1

Table 1 displays demographics and baseline characteristics after weighting by the inverse propensity score pertaining to the most severe group being compared and the corresponding standardized differences between the compared groups. All variables were well balanced among the comparison groups based on our predefined standardized difference criteria. Thus, after weighting by the relevant inverse propensity score when comparing groups on outcomes, there was no need to adjust for baseline variables. The raw summary statistics of demographics and baseline characteristics without the weighting are provided in Appendix 1.

No difference was found among the hypothyroid, treated, and euthyroid groups on the composite of in-hospital mortality or morbidity (all P values ≥0.30). The severity-weighted average relative effect across the outcome components was estimated as 1.14 (98.3% CI, 0.85–1.52) for hypothyroid versus treated, 1.07 (98.3% CI, 0.83–1.37) for hypothyroid versus euthyroid, and 0.94 (98.3% CI, 0.74–1.20) for treated versus euthyroid (Table 3). Our sensitivity analyses using a multiple propensity score method and a propensity score matching method provided the same conclusions and very similar effect estimates (Table 3).

Table 3

Table 3

The association with hypothyroidism and outcome was not consistent across the 3 components of the primary outcome (hypothyroid-by-outcome interaction, P = 0.05). Therefore, we also evaluated the associations among the hypothyroid, treated, and euthyroid groups for each of the 3 components of the composite outcome and found no associations (Table 4). Significant heterogeneity, especially with estimates in opposite directions as seen in Table 4, would suggest that the individual ORs be given more importance than the overall OR.30,31

Table 4

Table 4

We observed that 502 (63%) hypothyroid patients, 1043 (58%) treated patients, and 3186 (57%) euthyroid patients received vasopressor during surgery. After confounding adjustment, hypothyroid patients were slightly, but significantly, more likely to be given an intraoperative vasopressor than either treated (OR, 1.17; 99.2% CI, 1.01–1.36) or euthyroid (OR, 1.12; 99.2% CI, 1.02–1.24) patients (Table 5).

Table 5

Table 5

There were 51 (6%) hypothyroid, 54 (3%) treated, and 224 (4%) euthyroid patients who died in the hospital. Discharges for those patients were considered as nonevents and censored at the longest observed length of stay. The estimated median durations of hospitalization from the Kaplan-Meier curve were 7 days (Q1–Q3, 2–19 days) for hypothyroid patients, 3 days (Q1–Q3, 1–9 days) for treated patients, and 4 days (Q1–Q3, 1–12 days) for euthyroid patients (Fig. 2). After confounding adjustment, hypothyroid patients had slightly longer length of stay (i.e., were slightly but significantly less likely to be discharged at any given time postoperatively) than treated patients (hazard ratio, 0.92; 99.2% CI, 0.86–0.99). No difference was found either between hypothyroid and euthyroid patients or between treated and euthyroid patients after Bonferroni correction (Table 5).

Figure 2

Figure 2

Appendix 2 and 3 show average and minimum systolic and diastolic blood pressures during each of the following intraoperative periods: start case to induction, induction to tracheal intubation, intubation to incision, incision to closing, closing to emergence, and emergence to end of case. The 3 groups were descriptively similar on systolic and diastolic blood pressures during surgery.

Back to Top | Article Outline

DISCUSSION

Our large-scale analysis of hypothyroidism and major postoperative outcomes found that a severity-weighted composite of in-hospital mortality and cardiovascular and surgical wound complications did not differ among hypothyroid, treated, and euthyroid patients. Hypothyroid rats have decreased type IV collagen and hydroxyproline during the proliferative phase of wound healing, which suggests that the thyroid hormone is associated with the proliferation and secretion of fibroblasts.12 Consistent with this theory, Mehregan and Zamick32 demonstrated a beneficial effect of thyroid hormone on healing of deep dermal burns in rats, with better organization of collagen bundles, fewer retraction spaces, and smoother scars. However, we did not observe significant differences in the composite of mortality, cardiovascular morbidity, and surgical wound complication or infection morbidity or in each component considered separately. A limitation of our analysis is that we measured TSH concentrations rather than thyroid hormone concentrations that have a direct impact on the wound healing process; thus, it remains possible (and perhaps likely) that many of our patients whom we considered hypothyroid based on the increased TSH concentrations actually had normal T3 and T4 concentrations.

Echocardiography studies show abnormalities of both systolic function (prolonged pre-ejection period of systole and left ventricular ejection time) and diastolic function in hypothyroid patients, which correlated with serum thyroid hormone levels.33 Some have proposed that even subclinical hypothyroidism may have adverse effects on the cardiovascular system.10,11 Furthermore, hypothyroid patients exhibit reduced sensitivity to catecholamines,34 possibly explaining the increased requirement for vasopressor support in our hypothyroid patients. However, systolic and diastolic blood pressures were similar in hypothyroid and other patients, suggesting that the degree of hypotension was of marginal clinical importance. Although hypothyroid patients were more likely to be given intraoperative vasopressors than treated and euthyroid patients, the slight difference was not clinically important. Thus, our results indicate that hypothyroid surgical patients do not suffer clinically important cardiovascular disturbances.

The median duration of hospitalization was 7 days in hypothyroid patients, 3 days in treated patients, and 4 days in euthyroid patients, with hypothyroid patients being significantly less likely to be discharged at any given postoperative time than treated patients. Ladenson et al.21 reported gastrointestinal complications and delayed recovery from anesthesia in hypothyroid patients after cardiac surgery and postoperative neuropsychiatric complications in hypothyroid patients after noncardiac surgery. We did not observe significant differences in cardiovascular and wound complications/infection morbidities in hypothyroid and euthyroid patients; however, it remains possible that hypothyroid patients had other postoperative complications that were not included in our analysis. As expected, we did not observe significant differences in occurrences of in-hospital mortality and complications between our treated (hypothyroid diagnosis with normal TSH values) and euthyroid (no hypothyroid diagnosis with normal TSH values) groups. Our results are consistent with the general belief that corrected hypothyroidism does not augment mortality.

We categorized patients as being hypothyroid when they had increased TSH concentrations within 6 months of surgery. Any TSH tests obtained anywhere within the Cleveland Clinic system were available to us, but tests at outside facilities were not. Thus, we do not know to what extent that patients categorized as hypothyroid may have been tested at other institutions.

Thyroid hormone concentrations, which are necessary for optimal cardiac function35 and tissue healing,36 were not available for most patients. It is possible that T3 and T4 concentrations were normal in some of our hypothyroid patients, even though TSH concentrations were elevated. Complications were defined by International Classification of Diseases, 9th Revision; although coding is generally accurate, administrative data do not contain the detail and precision of clinical or research-specific data.

An advantage of our study is that we used a multivariate (multiple outcome components per patient) analysis that adjusts for and takes advantage of the correlations among outcome components, facilitates applying differential severity weighting for the outcome components, and typically improves power compared with the traditional methods including analyzing the outcomes as a collapsed composite or separately. However, there are several limitations inherent to a retrospective study. Surely there remains some degree of selection bias because patients without TSH measurements were not included in the study, and those may theoretically be healthier patients. As with all retrospective studies, we only controlled for the observed potential confounding. There might be important confounding factors that are unavailable in our electronic records. Further, this was a single-center study; results may differ in other settings or populations. This study had 90% power to detect ORs of ≥1.4 between the hypothyroid group and each of the treated and euthyroid groups. Therefore, the study could have been underpowered to detect slight but clinically significant differences in occurrence of primary outcomes among the groups.

In summary, hypothyroidism slightly increased the requirement of intraoperative vasopressors and prolonged hospitalization after noncardiac inpatient surgery. However, intraoperative arterial blood pressures and a composite of in-hospital death and other complications were no different between hypothyroid and other patients. Thus, it seems unnecessary to postpone elective surgery in patients with mild hypothyroidism.

Back to Top | Article Outline

Appendix 1. Demographics and Baseline Characteristics

Table

Table

Back to Top | Article Outline

Appendix 2.

Plot of means and SD of the (A) average of and (B) minimum of systolic blood pressure during the following intraoperative time periods: from start of case to induction, from induction to intubation, from intubation to incision, from incision to closing, from closing to emergence, and from emergence to end of case

Plot of means and SD of the (A) average of and (B) minimum of systolic blood pressure during the following intraoperative time periods: from start of case to induction, from induction to intubation, from intubation to incision, from incision to closing, from closing to emergence, and from emergence to end of case

Back to Top | Article Outline

Appendix 3.

Plot of means and SD of the (A) average of and (B) minimum of diastolic blood pressure during the following intraoperative time periods: from start of case to induction, from induction to intubation, from intubation to incision, from incision to closing, from closing to emergence, and from emergence to end of case

Plot of means and SD of the (A) average of and (B) minimum of diastolic blood pressure during the following intraoperative time periods: from start of case to induction, from induction to intubation, from intubation to incision, from incision to closing, from closing to emergence, and from emergence to end of case

Back to Top | Article Outline

DISCLOSURES

Name: Ryu Komatsu, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Ryu Komatsu has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Name: Jing You, MS.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Jing You has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Edward J. Mascha, PhD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Edward J. Mascha has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

Name: Daniel I. Sessler, MD.

Contribution: This author helped design the study and write the manuscript.

Attestation: Daniel I. Sessler reviewed the analysis of the data and approved the final manuscript.

Name: Yusuke Kasuya, MD.

Contribution: This author helped design the study, conduct the study, analyze the data, and write the manuscript.

Attestation: Yusuke Kasuya has seen the original study data, reviewed the analysis of the data, approved the final manuscript, and is the author responsible for archiving the study files.

Name: Alparslan Turan, MD.

Contribution: This author helped design the study, conduct the study, and write the manuscript.

Attestation: Alparslan Turan has seen the original study data, reviewed the analysis of the data, and approved the final manuscript.

This manuscript was handled by: Sorin J. Brull, MD, FCARCSI (Hon).

Back to Top | Article Outline

REFERENCES

1. Tunbridge WM, Evered DC, Hall R, Appleton D, Brewis M, Clark F, Evans JG, Young E, Bird T, Smith PA. The spectrum of thyroid disease in a community: the Whickham survey. Clin Endocrinol (Oxf). 1977;7:481–93
2. Nyström E, Bengtsson C, Lindquist O, Noppa H, Lindstedt G, Lundberg PA. Thyroid disease and high concentration of serum thyrotrophin in a population sample of women. A 4-year follow-up. Acta Med Scand. 1981;210:39–46
3. Falkenberg M, Kågedal B, Norr A. Screening of an elderly female population for hypo- and hyperthyroidism by use of a thyroid hormone panel. Acta Med Scand. 1983;214:361–5
4. Okamura K, Ueda K, Sone H, Ikenoue H, Hasuo Y, Sato K, Yoshinari M, Fujishima M. A sensitive thyroid stimulating hormone assay for screening of thyroid functional disorder in elderly Japanese. J Am Geriatr Soc. 1989;37:317–22
5. Sawin CT, Castelli WP, Hershman JM, McNamara P, Bacharach P. The aging thyroid. Thyroid deficiency in the Framingham Study. Arch Intern Med. 1985;145:1386–8
6. Parle JV, Franklyn JA, Cross KW, Jones SC, Sheppard MC. Prevalence and follow-up of abnormal thyrotrophin (TSH) concentrations in the elderly in the United Kingdom. Clin Endocrinol (Oxf). 1991;34:77–83
7. Sawin CT, Geller A, Hershman JM, Castelli W, Bacharach P. The aging thyroid. The use of thyroid hormone in older persons. JAMA. 1989;261:2653–5
8. Dilmann WHBevermann LE, Utiger RD. The cardiovascular system in hypothyroidism. In: The Thyroid. 19916th ed. Philadelphia JP Lippincott:1002–8
9. Fredlund BO, Olsson SB. Long QT interval and ventricular tachycardia of “torsade de pointe” type in hypothyroidism. Acta Med Scand. 1983;213:231–5
10. Biondi B, Fazio S, Palmieri EA, Carella C, Panza N, Cittadini A, Bonè F, Lombardi G, Saccà L. Left ventricular diastolic dysfunction in patients with subclinical hypothyroidism. J Clin Endocrinol Metab. 1999;84:2064–7
11. Ripoli A, Pingitore A, Favilli B, Bottoni A, Turchi S, Osman NF, De Marchi D, Lombardi M, L’Abbate A, Iervasi G. Does subclinical hypothyroidism affect cardiac pump performance? Evidence from a magnetic resonance imaging study. J Am Coll Cardiol. 2005;45:439–45
12. Natori J, Shimizu K, Nagahama M, Tanaka S. The influence of hypothyroidism on wound healing. An experimental study. Nihon Ika Daigaku Zasshi. 1999;66:176–80
13. Smith TJ, Bahn RS, Gorman CA. Connective tissue, glycosaminoglycans, and diseases of the thyroid. Endocr Rev. 1989;10:366–91
14. Thá Nassif AC, Hintz Greca F, Graf H, Domingues Repka JC, Nassif LS. Wound healing in colonic anastomosis in hypothyroidism. Eur Surg Res. 2009;42:209–15
15. Ekmektzoglou KA, Zografos GC. A concomitant review of the effects of diabetes mellitus and hypothyroidism in wound healing. World J Gastroenterol. 2006;12:2721–9
16. Grunfeld R, Kunselman A, Bustillo J, Juliano PJ. Wound complications in thyroxine-supplemented patients following foot and ankle surgery. Foot Ankle Int. 2011;32:38–46
17. Hanasono MM, Lin D, Wax MK, Rosenthal EL. Closure of laryngectomy defects in the age of chemoradiation therapy. Head Neck. 2012;34:580–8
18. Shermak MA, Chang D, Magnuson TH, Schweitzer MA. An outcomes analysis of patients undergoing body contouring surgery after massive weight loss. Plast Reconstr Surg. 2006;118:1026–31
19. Yerushalmy A, Reches A, Lessing JB, Schechner V, Carmeli Y, Grisaru D. Characteristics of microorganisms cultured from infected wounds post-hysterectomy. Eur J Obstet Gynecol Reprod Biol. 2008;141:169–72
20. Myerowitz PD, Kamienski RW, Swanson DK, Chopra PS, Berkoff HA, Kroncke GM, Rowe GG, Vander Ark CR, Dhanani SP. Diagnosis and management of the hypothyroid patient with chest pain. J Thorac Cardiovasc Surg. 1983;86:57–60
21. Ladenson PW, Levin AA, Ridgway EC, Daniels GH. Complications of surgery in hypothyroid patients. Am J Med. 1984;77:261–6
22. Park YJ, Yoon JW, Kim KI, Lee YJ, Kim KW, Choi SH, Lim S, Choi DJ, Park KH, Choh JH, Jang HC, Kim SY, Cho BY, Lim C. Subclinical hypothyroidism might increase the risk of transient atrial fibrillation after coronary artery bypass grafting. Ann Thorac Surg. 2009;87:1846–52
23. Drucker DJ, Burrow GN. Cardiovascular surgery in the hypothyroid patient. Arch Intern Med. 1985;145:1585–7
24. Jones TH, Hunter SM, Price A, Angelini GD. Should thyroid function be assessed before cardiopulmonary bypass operations? Ann Thorac Surg. 1994;58:434–6
25. Syed AU, El Watidy AF, Akhlaque NB, Wahba A, El Oakley RM, Imran K, Al Bukhari EA, Al Fagih MR. Coronary bypass surgery in patients on thyroxin replacement therapy. Asian Cardiovasc Thorac Ann. 2002;10:107–10
26. Weinberg AD, Brennan MD, Gorman CA, Marsh HM, O’Fallon WM. Outcome of anesthesia and surgery in hypothyroid patients. Arch Intern Med. 1983;143:893–7
27. Curtis LH, Hammill BG, Eisenstein EL, Kramer JM, Anstrom KJ. Using inverse probability-weighted estimators in comparative effectiveness analyses with observational databases. Med Care. 2007;45:S103–7
28. Austin PC. Balance diagnostics for comparing the distribution of baseline covariates between treatment groups in propensity-score matched samples. Stat Med. 2009;28:3083–107
29. Imbens GW. The role of the propensity score in estimating dose-response functions. Biometrika. 2000;87:706–10
30. Mascha EJ, Imrey PB. Factors affecting power of tests for multiple binary outcomes. Stat Med. 2010;29:2890–904
31. Mascha EJ, Sessler DI. Statistical grand rounds: design and analysis of studies with binary-event composite endpoints: guidelines for anesthesia research. Anesth Analg. 2011;112:1461–71
32. Mehregan AH, Zamick P. The effect of triiodothyronine in healing of deep dermal burns and marginal scars of skin grafts. A histologic study. J Cutan Pathol. 1974;1:113–6
33. Vora J, O’Malley BP, Petersen S, McCullough A, Rosenthal FD, Barnett DB. Reversible abnormalities of myocardial relaxation in hypothyroidism. J Clin Endocrinol Metab. 1985;61:269–72
34. Polikar R, Kennedy B, Maisel A, Ziegler M, Smith J, Dittrich H, Nicod P. Decreased adrenergic sensitivity in patients with hypothyroidism. J Am Coll Cardiol. 1990;15:94–8
35. Kahaly GJ, Dillmann WH. Thyroid hormone action in the heart. Endocr Rev. 2005;26:704–28
36. Davis FB, Mousa SA, O’Connor L, Mohamed S, Lin HY, Cao HJ, Davis PJ. Proangiogenic action of thyroid hormone is fibroblast growth factor-dependent and is initiated at the cell surface. Circ Res. 2004;94:1500–6
© 2015 International Anesthesia Research Society