Key Points
Question
Does an association exist between preoperative anemia and morbidity or mortality after thyroidectomy in adults with thyroid cancer?
Findings
This cross-sectional cohort study used the American College of Surgeons National Surgical Quality Improvement Program database to identify 24 912 patients with thyroid cancer who underwent thyroidectomy. Compared with patients without preoperative anemia, those with preoperative anemia had an increase in postoperative 30-day overall morbidity (1.7-fold) and mortality (3-fold).
Meaning
For adult patients with thyroid cancer, optimization of anemia prior to thyroidectomy may be warranted.
Abstract
Importance
Despite the ease of preoperative anemia diagnosis and the availability of treatment options, the morbidity and mortality associated with this condition remain unacceptably high, and the literature describing the association of preoperative anemia with postoperative outcomes following thyroid surgery in patients with thyroid cancer remain sparse. Reporting outcomes in this patient population may help to facilitate preoperative optimization.
Objective
To assess whether an association exists between preoperative anemia and outcomes following thyroid surgery in patients with thyroid cancer.
Design, Setting, and Participants
This retrospective, cross-sectional, cohort study used the American College of Surgeons National Surgical Quality Improvement Program database to identify 32 166 patients between 2007 and 2016 with Current Procedural Terminology codes for thyroid surgery and with the International Classification of Diseases, Ninth Revision code of malignant thyroid cancer.
Exposures
Preoperative anemia as defined using the World Health Organization criteria of hematocrit less than 36% in nonpregnant females and less than 39% in males.
Main Outcomes and Measures
Multivariable logistic regression analysis was conducted to assess the association of preoperative anemia with the following 30-day postoperative outcomes: pulmonary, infectious, and cardiac complications, overall and serious morbidity (surgical site infection and medical complications), prolonged hospital length of stay (≥75th percentile for the cohort), and mortality.
Results
Among the 24 912 patients with thyroid cancer who underwent thyroidectomy included in the final analysis, the median (interquartile range) age was 51 (40-62) years and the majority were women (18 705 [75.1%]). The prevalence of preoperative anemia was 12.5% (n = 3108). Within the overall study population, hypertension (9242 patients [37.1%]) followed by active smoking (2992 patients [12.0%]) were the most prevalent comorbidities. The unadjusted odds of anemia vs no anemia were significantly higher for every 10-year increase in age (odds ratio [OR], 1.10; 95% CI, 1.08-1.13) and for black vs white patients (OR, 2.82; 95% CI, 2.51-3.17). The adjusted odds of postoperative overall morbidity (OR, 1.68; 95% CI, 1.29-2.17), mortality (OR, 3.36; 95% CI, 1.37-8.28), and pulmonary (OR, 2.36; 95% CI, 1.65-3.34) and infectious (OR, 1.62; 95% CI, 1.12-2.29) complications were higher in patients with preoperative anemia than in those without preoperative anemia.
Conclusions and Relevance
The findings from this study suggest that preoperative anemia may not only be associated with racial differences and a higher comorbidity burden but may also increase the likelihood of postoperative morbidity and mortality. These results may provide a basis for further risk reduction strategies and preoperative optimization.
This cross-sectional cohort study uses the American College of Surgeons National Surgical Quality Improvement Program database to evaluate whether an association exists between preoperative anemia and postoperative 30-day morbidity and mortality outcomes following thyroidectomy in adult patients with thyroid cancer.
Introduction
Anemia is the most common hematologic condition in the preoperative setting and is often caused by underlying comorbidities.1,2 The association of preoperative anemia with postoperative morbidity and mortality has been well described across various surgical specialties.3,4,5,6,7 Despite the ease of diagnosis and availability of treatment options, the morbidity and mortality associated with preoperative anemia remain unacceptably high. Although the association of preoperative anemia with postoperative outcomes has been well described within diverse populations, the literature on the association of preoperative anemia with outcomes following thyroid surgery in patients with thyroid cancer remains sparse. In the United States, the lifetime risk of thyroid cancer is only 1.2%; however, the incidence has steadily increased every year, which may be attributable to advanced detection technologies.8 The National Cancer Institute stated that the incidence and mortality rates of thyroid cancer increased from 2005 to 2014 by an average of 3.8% and 0.7% per year, respectively.9
Not surprisingly, preoperative anemia is a strong indicator for perioperative blood transfusion.3,10,11 Moreover, several studies have shown that perioperative blood transfusions are associated with poor prognosis and complications following surgery for head and neck cancer.10,12,13 Although studies in head and neck cancer have shown an association of perioperative blood transfusion with postoperative outcomes, the association of preoperative anemia with outcomes following thyroidectomy remains poorly defined. In a multivariable Surveillance, Epidemiology, and End Result database analysis, older age, a Charlson-Deyo comorbidity score of 2 or higher, and regional or distant disease were predictors of postoperative morbidity and mortality following surgery for thyroid cancer.14 To our knowledge, no study has assessed the relationship between preoperative anemia and postoperative 30-day morbidity and mortality in the context of thyroid surgery among patients with thyroid cancer. Therefore, we aimed to use the American College of Surgeons National Surgical Quality Improvement Program (NSQIP) database to identify whether such an association exists in adult patients with thyroid cancer following thyroidectomy. We hypothesize that among patients with thyroid cancer who undergo thyroidectomy, preoperative anemia is an independent risk factor associated with postoperative 30-day morbidity and mortality.
Methods
Data Collection and Study Population
Data were obtained from the publicly available NSQIP multicenter data set for the years 2007 to 2016.15 This database contains several variables, including preoperative risk factors and 30-day postoperative outcomes, from more than 200 participating hospitals for patients undergoing surgical procedures. A site-specific surgical clinical reviewer collects the data from medical charts. The NSQIP database undergoes an 8-day cycle sampling process, developed to ensure that cases have an equal chance of being selected from each day of the week. This study was exempt from University of California, San Diego, Institutional Review Board approval. The NSQIP database is deidentified and therefore meets the criteria of the Health Insurance Portability and Accountability Act to protect personal information; thus, this study was exempted by the same institutional review board from the need to obtain informed patient consent.
We used the following Current Procedural Terminology codes to identify patients who underwent thyroid surgical procedures: 60210 (unilateral partial thyroid lobectomy), 60212 (unilateral partial thyroid lobectomy with contralateral subtotal lobectomy), 60220 (unilateral total thyroid lobectomy), 60225 (unilateral total thyroid lobectomy with contralateral subtotal lobectomy), 60240 (total thyroidectomy), 60252 (thyroidectomy for malignancy with limited neck dissection), 60254 (thyroidectomy for malignancy with radical neck dissection), 60260 (thyroidectomy for removal of all remaining tissue), 60270 (thyroidectomy, including substernal thyroid, transthoracic approach), and 60271 (thyroidectomy, including substernal thyroid, cervical approach).16 We used the International Classification of Diseases, Ninth Revision code 193.0 to identify patients with malignant thyroid cancer.17 This retrospective analysis adheres to the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline for cross-sectional studies.18
In this retrospective cohort study, our primary exposure variable was preoperative anemia, defined based on the World Health Organization definitions of hematocrit less than 39% in males and less than 36% in nonpregnant females (to convert to a proportion of 1, multiply by 0.01). Data for potential confounding demographic variables were collected and included sex (male or female), race (white, black, Asian, Native American or Alaska Native, Native Hawaiian or Pacific Islander, or unknown), age, and body mass index (calculated as weight in kilograms divided by height in meters squared) and classified using World Health Organization criteria (underweight, <18.5; normal weight, 18.5-24.99; overweight, ≥25 and ≤29.99; or obese, ≥30). We also adjusted for the following preoperative comorbidities: functional status, steroid use (ie, requiring oral or parenteral corticosteroids 30 days prior to the surgical procedure), surgery type (ie, inpatient vs outpatient), history of congestive of heart failure, bleeding disorder, disseminated cancer, smoking history (ie, smoked cigarettes within the year prior to surgery), American Society of Anesthesiologists Physical Status (ASA PS) classification, history of chronic obstructive pulmonary disease, chemotherapy, radiotherapy, and perioperative red blood cell transfusion. Finally, we adjusted for type of thyroidectomy.
We evaluated the association of preoperative anemia on the following 30-day postoperative outcomes: overall morbidity, serious morbidity, pulmonary complications, infectious complications, renal complications, vascular complications, neurologic complications, wound dehiscence (ie, complete disruption of surgical site compromising the integrity of procedure), cardiac complications, prolonged hospital length of stay (≥75th percentile for the cohort), reoperation (ie, return to the operating room for any reason within 30 days of thyroid surgery), and mortality. In brief, we defined overall morbidity as sepsis, septic shock, deep venous thrombosis requiring therapy, pulmonary embolism, cerebrovascular accident, myocardial infarction, cardiac arrest requiring cardiopulmonary resuscitation, progressive renal insufficiency, acute renal failure, urinary tract infection, reintubation, ventilator dependence, pneumonia, wound infection, wound disruption, organ space surgical site infection, or superficial surgical site infection. We defined serious morbidity as organ space surgical site infection, wound disruption, cerebrovascular accident, myocardial infarction, cardiac arrest requiring cardiopulmonary resuscitation, pulmonary embolism, ventilator dependence, progressive renal insufficiency, acute renal failure, sepsis, or septic shock. Pulmonary complications were defined as pneumonia, unplanned reintubation, or prolonged ventilator use (ie, more than 48 hours). Infectious complications were defined as superficial incisional surgical site infection, wound infection, organ space infection, sepsis, or septic shock. Renal complications were defined as urinary tract infection, progressive renal insufficiency, or acute renal failure. Pulmonary embolism and deep venous thrombosis defined vascular complications. Cardiac complications were defined as myocardial infarction and cardiopulmonary resuscitation. The NSQIP participant user data file contains detailed information of all the variables used in this study.15
Statistical Analysis
The software environment for statistical computing, R (version 3.3.2), was used to perform all statistical analysis. We used the tableone package and library in R to generate frequency tables of anemia cohorts. A univariable logistic regression analysis was performed to evaluate the association of demographic variables, preoperative comorbidities, intraoperative variables, and postoperative outcomes with anemia. A multivariable logistic regression was performed to evaluate the association of preoperative anemia with outcomes of interest while also controlling for demographic variables, preoperative comorbidities, and intraoperative variables (discussed above). For the final model, an interaction term for preoperative anemia and perioperative transfusion was forced into the model. This interaction term was included as a secondary aim of our analysis to assess the association between preoperative anemia and perioperative transfusion needs with surgical outcomes (ie, overall and serious morbidity, prolonged hospital length of stay, mortality, and pulmonary, infectious, and cardiac complications). A 2-tailed significance level was set at P ≤ .05. The odds ratio (OR) with associated 95% CI is reported for each covariate.
Results
Of the 32 166 patients in the NSQIP database with thyroid cancer who underwent thyroid surgery from 2007 to 2016, 24 912 were included in the analysis. We excluded 7254 cases (22.5%) from the analysis owing to missing values; all of these cases were missing preoperative hematocrit values. The prevalence of preoperative anemia was 12.5% (n = 3108). Table 1 gives the distribution of patient demographic and clinical characteristics. Among the overall study population, the median (interquartile range) age was 51 (40-62) years. Women (18 705, 75.1%) represented the majority of the study population. Approximately one-third of the patients (n = 7365) had an ASA PS classification of 3 or higher (a patient with at least severe systemic disease). Most patients underwent total thyroidectomy (10 765, 43.2%) or total thyroidectomy with neck dissection (7378, 29.6%). The unadjusted odds of preoperative anemia were significantly higher for several demographic variables and preoperative clinical conditions (Table 1). The unadjusted odds of anemia vs no anemia were significantly higher for every 10-year increase in age (OR, 1.10; 95% CI, 1.08-1.13) and for black vs white patients (OR, 2.82; 95% CI, 2.51-3.17). The unadjusted odds of preoperative anemia were significantly higher for patients with overall morbidity (OR, 2.63; 95% CI, 2.10-3.26) or serious morbidity (OR, 3.59; 95% CI, 2.60-4.89), pulmonary (OR, 4.10; 95% CI, 3.03-5.51), infectious (OR, 2.30; 95% CI, 1.67-3.12), or cardiac (OR, 4.97; 95% CI, 2.31-10.33) complications, prolonged hospital length of stay of 1 or more days (OR, 1.36; 95% CI, 1.20-1.54), or mortality (OR, 8.38; 95% CI, 4.30-16.51) (Table 2).
Table 1. Distribution of Demographic and Clinical Characteristics Among Preoperative Anemia Cohorts Following Thyroid Surgery in Patients With Thyroid Cancer.
| Characteristic | Patients, No. (%) | OR (95% CI) | ||
|---|---|---|---|---|
| Overall Study Population | Without Anemiaa | With Anemiaa | ||
| Total, No. | 24 912 | 21 804 | 3108 | |
| Race | ||||
| White | 18 537 (74.4) | 16 418 (75.3) | 2119 (68.2) | 1 [Reference] |
| Black | 1683 (6.8) | 1234 (5.7) | 449 (14.4) | 2.82 (2.51-3.17) |
| Asian | 1305 (5.2) | 1158 (5.3) | 147 (4.7) | 0.98 (0.82-1.17) |
| Native American or Alaska Native | 96 (0.4) | 79 (0.4) | 17 (0.5) | 1.67 (0.95-2.75) |
| Native Hawaiian or Pacific Islander | 81 (0.3) | 74 (0.3) | 7 (0.2) | 0.73 (0.31-1.48) |
| Unknown | 3210 (12.9) | 2841 (13.0) | 369 (11.9) | 1.01 (0.89-1.13) |
| BMI | ||||
| 18.5 to 24.99 | 5665 (22.7) | 4917 (22.6) | 748 (24.1) | 1 [Reference] |
| <18.5 | 250 (1.0) | 209 (1.0) | 41 (1.3) | 1.29 (0.90-1.80) |
| ≥25 to ≤29.99 | 6674 (26.8) | 5910 (27.1) | 764 (24.6) | 0.85 (0.76-0.95) |
| ≥30 | 12 218 (49.0) | 10 673 (48.9) | 1545 (49.7) | 0.95 (0.87-1.05) |
| Unknown | 105 (0.4) | 95 (0.4) | 10 (0.3) | 0.69 (0.34-1.27) |
| Age, median (IQR), y | 51 (40-62) | 51 (40-62) | 52 (41-66) | 1.10 (1.08-1.13)b |
| Sex | ||||
| Male | 6194 (24.9) | 5403 (24.8) | 791 (25.5) | 1 [Reference] |
| Female | 18 705 (75.1) | 16 388 (75.2) | 2317 (74.5) | 0.97 (0.89-1.05) |
| Surgery status | ||||
| Outpatient | 12 333 (49.5) | 10 956 (50.2) | 1377 (44.3) | 1 [Reference] |
| Inpatient | 12 579 (50.5) | 10 848 (49.8) | 1731 (55.7) | 1.27 (1.18-1.37) |
| Dyspnea | ||||
| None | 23 557 (94.6) | 20 704 (95.0) | 2853 (91.8) | 1 [Reference] |
| Moderate exertion | 1253 (5.0) | 1026 (4.7) | 227 (7.3) | 1.61 (1.38-1.86) |
| At rest | 102 (0.4) | 74 (0.3) | 28 (0.9) | 2.75 (1.75-4.20) |
| COPD | 460 (1.8) | 370 (1.7) | 90 (2.9) | 1.73 (1.36-2.17) |
| CHF | 44 (0.2) | 26 (0.1) | 18 (0.6) | 4.88 (2.63-8.85) |
| Transfusion | 9 (0.0) | 1 (0.0) | 8 (0.3) | 56.27 (10.32-1044.03) |
| Bleeding disorder | 348 (1.4) | 264 (1.2) | 84 (2.7) | 2.27 (1.76-2.89) |
| Functional status | ||||
| Independent | 24 706 (99.4) | 21 668 (99.6) | 3038 (98.3) | 1 [Reference] |
| Partial dependence | 117 (0.5) | 74 (0.3) | 43 (1.4) | 4.14 (2.82-6.02) |
| Total dependence | 20 (0.1) | 9 (0.0) | 11 (0.4) | 8.72 (3.61-21.64) |
| Unknown | 69 (0.3) | 53 (0.2) | 16 (0.5) | 2.15 (1.19-3.68) |
| Diabetes | ||||
| None | 21 857 (87.7) | 19 383 (88.9) | 2474 (79.6) | 1 [Reference] |
| Insulin nondependent | 2104 (8.4) | 1722 (7.9) | 382 (12.3) | 1.74 (1.54-1.95) |
| Insulin dependent | 951 (3.8) | 699 (3.2) | 252 (8.1) | 2.82 (2.43-3.28) |
| Active smoker | 2991 (12.0) | 2672 (12.3) | 319 (10.3) | 0.82 (0.72-0.92) |
| Alcohol use | ||||
| None | 10 336 (41.5) | 8964 (41.1) | 1372 (44.1) | 1 [Reference] |
| Yes | 127 (0.5) | 109 (0.5) | 18 (0.6) | 1.08 (0.63-1.74) |
| Unknown history | 14 449 (58.0) | 12 731 (58.4) | 1718 (55.3) | 0.88 (0.82-0.95) |
| Disseminated cancer | 520 (2.1) | 359 (1.6) | 161 (5.2) | 3.26 (2.69-3.94) |
| Long-term steroid use | 507 (2.0) | 389 (1.8) | 118 (3.8) | 2.17 (1.76-2.67) |
| Renal dialysis | 91 (0.4) | 32 (0.1) | 59 (1.9) | 13.17 (8.61-20.50) |
| Hypertension | 9242 (37.1) | 7741 (35.5) | 1501 (48.3) | 1.70 (1.57-1.83) |
| Emergency surgery | 54 (0.2) | 38 (0.2) | 16 (0.5) | 2.96 (1.61-5.22) |
| Ventilator dependence | 18 (0.1) | 1 (0.0) | 17 (0.5) | 119.9 (24.63-2161.18) |
| Weight loss | 141 (0.6) | 99 (0.5) | 42 (1.4) | 3.00 (2.07-4.28) |
| History of TIA | ||||
| None | 10 296 (41.3) | 8926 (40.9) | 1370 (44.1) | 1 [Reference] |
| Yes | 157 (0.6) | 137 (0.6) | 20 (0.6) | 0.95 (0.58-1.49) |
| Unknown history | 14 459 (58.0) | 12 741 (58.4) | 1718 (55.3) | 0.88 (0.81-0.95) |
| CVA | ||||
| None | 10 360 (41.6) | 8990 (41.2) | 1370 (44.1) | 1 [Reference] |
| Yes | 93 (0.4) | 73 (0.3) | 20 (0.6) | 1.80 (1.06-2.90) |
| Unknown history | 14 459 (58.0) | 12 741 (58.4) | 1718 (55.3) | 0.88 (0.82-0.95) |
| CVANO | ||||
| None | 10 364 (41.6) | 8991 (41.2) | 1373 (44.2) | 1 [Reference] |
| Yes | 85 (0.3) | 68 (0.3) | 17 (0.5) | 1.64 (0.93-2.73) |
| Unknown history | 14 463 (58.1) | 12 745 (58.5) | 1718 (55.3) | 0.88 (0.82-0.95) |
| Prior surgery | ||||
| None | 10 238 (48.2) | 8894 (48.0) | 1344 (49.5) | 1 [Reference] |
| Yes | 207 (1.0) | 164 (0.9) | 43 (1.6) | 1.74 (1.22-2.42) |
| Unknown history | 14 467 (58.1) | 12 746 (58.5) | 1721 (55.4) | 0.93 (0.86-1.01) |
| Chemotherapy | ||||
| None | 10 410 (49.0) | 9046 (48.8) | 1364 (50.2) | 1 [Reference] |
| Yes | 43 (0.2) | 17 (0.1) | 26 (1.0) | 10.14 (5.53-19.08) |
| Unknown history | 14 459 (58.0) | 12 741 (58.4) | 1718 (55.3) | 0.93 (0.86-1.01) |
| Radiotherapy | ||||
| None | 10 407 (49.0) | 9033 (48.7) | 1374 (50.6) | 1 [Reference] |
| Yes | 34 (0.2) | 21 (0.1) | 13 (0.5) | 4.07 (1.98-8.05) |
| Unknown history | 14 471 (58.1) | 12 750 (58.5) | 1721 (55.4) | 0.92 (0.85-1.00) |
| PCI | ||||
| None | 10 226 (48.1) | 8903 (48.0) | 1323 (48.7) | 1 [Reference] |
| Yes | 227 (1.1) | 160 (0.9) | 67 (2.5) | 2.82 (2.09-3.75) |
| Unknown history | 14 459 (58.0) | 12 741 (58.4) | 1718 (55.3) | 0.94 (0.87-1.02) |
| Previous cardiac surgery | ||||
| None | 10 271 (48.3) | 8934 (48.2) | 1337 (49.2) | 1 [Reference] |
| Yes | 182 (0.9) | 129 (0.7) | 53 (2.0) | 2.75 (1.97-3.78) |
| Unknown history | 14 459 (58.0) | 12 741 (58.4) | 1718 (55.3) | 0.94 (0.86-1.01) |
| Peripheral vascular disease | ||||
| None | 10 418 (41.8) | 9037 (41.4) | 1381 (44.4) | 1 [Reference] |
| Yes | 35 (0.1) | 26 (0.1) | 9 (0.3) | 2.27 (1.00-4.67) |
| Unknown history | 14 459 (58.0) | 12 741 (58.4) | 1718 (55.3) | 0.88 (0.82-0.95) |
| ASA PS classificationc | ||||
| ≤2 | 17 502 (70.3) | 15 793 (72.4) | 1709 (55.0) | 1 [Reference] |
| ≥3 | 7365 (29.6) | 5974 (27.4) | 1391 (44.8) | 2.15 (1.99-2.32) |
| Unknown | 45 (0.2) | 37 (0.2) | 8 (0.3) | 2.00 (0.86-4.08) |
| Work RVU, median (IQR) | 16.18 (15.04-21.88) | 16.18 (15.04-21.88) | 16.18 (15.04-21.88) | 1.00 (0.99-1.00) |
| Case duration, median (IQR), h | 1.88 (1.35-2.63) | 1.88 (1.33-2.63) | 1.93 (1.40-2.73) | 1.04 (1.01-1.06) |
| Thyroid surgery | ||||
| Partial thyroid lobectomy | 767 (3.1) | 672 (3.1) | 95 (3.1) | 1 [Reference] |
| Total thyroid lobectomy | 3990 (16.0) | 3491 (16.0) | 499 (16.1) | 1.01 (0.80-1.28) |
| Total thyroidectomy | 10 765 (43.2) | 9408 (43.1) | 1357 (43.7) | 1.02 (0.82-1.28) |
| Total thyroidectomy with neck dissection | 7378 (29.6) | 6510 (29.9) | 868 (27.9) | 0.94 (0.76-1.19) |
| Completion thyroidectomy | 1643 (6.6) | 1415 (6.5) | 228 (7.3) | 1.14 (0.88-1.48) |
| Thyroidectomy with substernal thyroid removal | 369 (1.5) | 308 (1.4) | 61 (2.0) | 1.40 (0.98-1.98) |
Abbreviations: ASA PS, American Society of Anesthesiologists Physical Status; BMI, body mass index (calculated as weight in kilograms divided by height in meters squared); CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; CVA, cerebrovascular accident without neurologic deficit; CVANO, cerebrovascular accident with neurologic deficit; IQR, interquartile range; OR, odds ratio; PCI, percutaneous coronary intervention; RVU, relative value unit; TIA, transient ischemic attack.
Anemia defined based on World Health Organization criteria: hematocrit in males less than 39%; hematocrit in nonpregnant females less than 36% (to convert hematocrit to a proportion of 1, multiple by 0.01).
Odds ratio for age represents 10-y increases in age.
A score between 1 to 6 represents the patient’s health status, which helps to predict operative risk. The higher the score the greater the disease burden and operative risk.
Table 2. Distribution of Postoperative Outcomes Among Preoperative Anemia Cohorts Following Thyroid Surgery in Patients With Thyroid Cancer.
| Outcomea | Participants, No. (%) | OR (95% CI) | ||
|---|---|---|---|---|
| Overall Study Population | Without Anemia | With Anemia | ||
| Total, No. | 24 912 | 21 804 | 3108 | |
| Morbidity | ||||
| Overall | 418 (1.7) | 306 (1.4) | 112 (3.6) | 2.63 (2.10-3.26) |
| Serious | 176 (0.7) | 117 (0.5) | 59 (1.9) | 3.59 (2.60-4.89) |
| Pulmonary | 189 (0.8) | 120 (0.6) | 69 (2.2) | 4.10 (3.03-5.51) |
| Infectious | 216 (0.9) | 163 (0.7) | 53 (1.7) | 2.30 (1.67-3.12) |
| Renal | 10 (0.0) | 8 (0.0) | 2 (0.1) | 1.75 (0.26-7.00) |
| Vascular | 32 (0.1) | 24 (0.1) | 8 (0.3) | 2.34 (0.98-5.00) |
| Neurologic | 13 (0.1) | 9 (0.0) | 4 (0.1) | 3.12 (0.85-9.59) |
| Wound dehiscence | 8 (0.0) | 6 (0.0) | 2 (0.1) | 2.34 (0.34-10.16) |
| Cardiac | 29 (0.1) | 17 (0.1) | 12 (0.4) | 4.97 (2.31-10.33) |
| Prolonged hospital stay, db | ||||
| <1 | 3122 (12.5) | 2816 (12.9) | 306 (9.8) | 1 [Reference] |
| ≥1 | 21 787 (87.5) | 18 985 (87.1) | 2802 (90.2) | 1.36 (1.20-1.54) |
| Unknown | 3 (0.0) | 3 (0.0) | 0 (0.0) | c |
| Total hospital stay, No., mean (SD), d | 1.37 (4.29) | 1.29 (4.24) | 1.96 (4.58) | 1.04 (1.03-1.06) |
| Reoperation | 597 (2.4) | 522 (2.4) | 75 (2.4) | 1.01 (0.78-1.28) |
| Death | 35 (0.1) | 16 (0.1) | 19 (0.6) | 8.38 (4.30-16.51) |
Abbreviation: OR, odds ratio.
Definitions: morbidity, 30-d complications; pulmonary, 30-d ventilator dependence, pneumonia, or reintubation; infectious, superficial incisional surgical site infection, organ space surgical site infection, sepsis, or septic shock; renal, urinary tract infection, progressive renal insufficiency, or acute renal failure; vascular, pulmonary embolism, or deep venous thrombosis; neurologic, cerebrovascular accident; cardiac, myocardial infarction or cardiac arrest requiring cardiopulmonary resuscitation.
Defined as the 75th percentile for the cohort.
Low sample size and unable to calculate OR.
Table 3 gives the results of multivariable logistic regression analysis. All models were adjusted for race, body mass index, age, sex, chronic obstructive pulmonary disease, congestive heart failure, preoperative transfusion, bleeding disorder, functional status, active smoking, disseminated cancer, steroid use, ASA PS classification, chemotherapy, radiotherapy, work relative value units, thyroid surgery, and surgery type (ie, inpatient vs outpatient). The adjusted odds of postoperative overall morbidity (OR, 1.68; 95% CI, 1.29-2.17), mortality (OR, 3.36; 95% CI, 1.37-8.28), or pulmonary (OR, 2.36; 95% CI, 1.65-3.34), or infectious (OR, 1.62; 95% CI, 1.12-2.29) complications were higher in patients with rather than without preoperative anemia. After adjustment, postoperative cardiac complications were twice as likely in patients with preoperative anemia (OR, 2.08; 95% CI, 0.75-5.37), and prolonged hospital length of stay of 1 or more days (OR, 1.14; 95% CI, 0.98-1.33) was also more likely in patients with anemia although the imprecision of the estimates as defined by the width of the 95% CIs and inclusion of the null value prevents definitive conclusions. The interaction between preoperative anemia and perioperative red blood cell transfusion for all surgical outcomes (ie, overall and serious morbidity, prolonged hospital length of stay, mortality, and pulmonary, infectious, and cardiac complications) was nonsignificant (all P > .05).
Table 3. Multivariable Analysis of Preoperative Anemia and 30-Day Postoperative Outcomesa.
| Outcome | Odds Ratio (95% CI)a |
|---|---|
| Morbidity | |
| Overall | 1.68 (1.29-2.17) |
| Serious | 2.17 (1.48-3.14) |
| Pulmonary | 2.36 (1.65-3.34) |
| Infectious | 1.62 (1.12-2.29) |
| Cardiac | 2.08 (0.75-5.37) |
| Prolonged hospital stay (≥1 d) | 1.14 (0.98 -1.33) |
| Death | 3.36 (1.37-8.28) |
Anemia reference group is no anemia.
Discussion
In this cross-sectional analysis of NSQIP data, we showed a 12.5% prevalence of preoperative anemia among patients with thyroid cancer who underwent thyroidectomy. After adjusting for demographic factors, comorbidities, and perioperative blood transfusion, our multivariable analysis identified increased odds of overall and serious morbidity, pulmonary and infectious complications, and mortality in patients with anemia. Furthermore, the association of preoperative anemia with postoperative outcomes for patients receiving perioperative transfusion vs those who did was not significant. To our knowledge, this is the first study using a surgical outcomes database to identify the association of preoperative anemia and 30-day postoperative adverse events in this surgical population. The prevalence of preoperative anemia was high despite the ease of preoperative diagnosis and several treatment options. Further preoperative screening, anemia optimization, and patient-centered education are needed.
The 4 main types of thyroid cancer, that is, papillary, follicular, medullary, and anaplastic carcinoma, represent 96.5% of all thyroid cancers.8,9 Per the National Cancer Institute, since 1975 the incidence of thyroid cancer has been on the rise.9 Data from the National Cancer Institute suggest that the incidence rose from 4.9 per 100 000 persons in 1975 to 14.3 per 100 000 persons in 2009.9 It is well established that thyroid cancer is not only 2 to 3 times more common in women but that the incidence is also higher among Asian and white individuals than in black and Native American individuals. The standard of care for thyroid cancer is thyroidectomy, and population-based data have been used to report perioperative complications. In a retrospective study using Surveillance, Epidemiology, and End Result data, of the patients who underwent thyroid surgery, 6.5% developed postoperative complications (ie, pneumonia, emergency intubation, tracheostomy, myocardial infarction, pulmonary embolism, deep venous thrombosis, fever, infection, or hemorrhage or hematoma) and 12.3% developed thyroid surgery–specific complications (ie, hypocalcemia or hypoparathyroidism and vocal cord or vocal fold paralysis).14 Despite an overall 5-year survival rate of 98.1%, general- and thyroid surgery–specific complication rates remain high and further preoperative rehabilitation is warranted.9
Preoperative anemia was traditionally treated with allogenic transfusion; however, blood supply shortages coupled with the increased risk of morbidity and mortality associated with transfusion has called for alternative management strategies. In a retrospective observational study assessing the association of preoperative anemia and transfusion in approximately 1 million patients who underwent noncardiac surgical procedures, it was shown that patients without anemia who did not receive a transfusion had a 0.2% mortality rate, whereas patients with preoperative anemia who received a transfusion had an 5.4% mortality rate.19 Considering the risks associated with perioperative transfusion, patient-centered management strategies have emerged to correct perioperative anemia, such as the 3-pronged approach of optimizing red blood cell mass, reducing perioperative blood loss, and increasing the physiological tolerance of preoperative anemia.20 Autologous transfusions, iron replacement therapy, and erythropoiesis-stimulating agents have more recently been considered in the treatment of preoperative anemia.21 Considering the negative contribution of transfusion to outcomes, prior to surgery, the severity and underlying cause of anemia should be evaluated and correction should be considered when possible; however, potential corrections must be weighed against the risks associated with delaying surgery for individuals with cancer.
The 30-day morbidity and mortality are significantly increased in patients with preoperative anemia in head and neck surgery. Rutledge et al22 conducted a retrospective review of 4648 patients who underwent total laryngectomy and found that anemia is associated with higher postoperative complication rates. In their evaluation of 971 total laryngectomy cases, Lebo et al23 showed that preoperative transfusion is associated with pharyngocutaneous fistula. Such findings demonstrate that preoperative anemia, reflected in the need for preoperative transfusion, likely represents either the degree of nutritional impairment or the comorbidity burden, both of which are also independently associated with impaired wound healing and postoperative morbidity. Studies across surgical disciplines have shown preoperative anemia to be predictive of increased morbidity and mortality.24,25,26,27,28,29 Phan et al30 conducted a retrospective study evaluating preoperative anemia with outcomes in 473 patients undergoing cervical spine surgery. They found that preoperative anemia is associated with higher comorbidity burden (ie, diabetes, dependent functional status, or an ASA PS classification ≥3), postoperative morbidity (ie, pulmonary complications, blood transfusion, reoperation, readmission, or extended hospital length of stay), and mortality. Other studies in patients who underwent cardiac surgery have shown preoperative anemia to be associated with higher rates of inpatient death, longer intensive care unit stays, and extended postoperative hospital length of stay.31 Similarly, in their evaluation of preoperative anemia with postoperative outcomes, Saager et al32 have shown that patients with preoperative anemia undergoing noncardiac surgery are 24% more likely to experience 30-day mortality. Often, patients who have received a diagnosis of cancer are unable to undergo optimal preoperative rehabilitation. The benefits of optimizing anemia vs the risks associated with waiting to remove the cancer need to be weighed.
Limitations
Although the NSQIP database is an excellent resource for surgical outcome studies, there are limitations to the present NSQIP data analysis. The NSQIP database does not include several important perioperative variables that would enable more robust evaluation of preoperative anemia and 30-day postoperative outcomes, such as the date of preoperative anemia diagnosis, volume of red blood cells transfused, perioperative fluid administration, and administration of iron replacement therapy or erythropoiesis-stimulating agents. The NSQIP data set is a large prospective outcome-oriented database in which granular information may not be obtained; therefore, we were unable to determine whether outcomes evaluated in this study were secondary to surgery- and anesthetic-specific causes or to the patient’s medical comorbidity burden. Because the NSQIP database primarily provides surgical outcomes, other important clinical data relevant to thyroid cancer are not available, such as cancer type and stage. The retrospective nature of the study design introduces limitations and may compromise the generalizability of our results. Registry with the NSQIP database is voluntary, with program demands that may limit the participation of hospitals, ultimately reducing the generalizability of the data. The NSQIP database does not include information on other comorbidities that may influence outcomes assessed in our study. Finally, the NSQIP database does not include surgeon or anesthesiologist clinical experience and hospital volume.
Conclusions
To our knowledge, this is the first population-based study to evaluate the association of preoperative anemia and 30-day postoperative outcomes in patients with thyroid cancer who underwent thyroid surgery. After controlling for potential confounders, we showed that preoperative anemia was associated with several postoperative outcomes (ie, morbidity, mortality, and pulmonary and infectious complications). These results provide a strong basis for further risk reduction strategies and program implementation.
References
- 1.Elhenawy AM, Meyer SR, Bagshaw SM, MacArthur RG, Carroll LJ. Role of preoperative intravenous iron therapy to correct anemia before major surgery: study protocol for systematic review and meta-analysis. Syst Rev. 2015;4:29. doi: 10.1186/s13643-015-0016-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wu WC, Schifftner TL, Henderson WG, et al. Preoperative hematocrit levels and postoperative outcomes in older patients undergoing noncardiac surgery. JAMA. 2007;297(22):2481-2488. doi: 10.1001/jama.297.22.2481 [DOI] [PubMed] [Google Scholar]
- 3.LaPar DJ, Hawkins RB, McMurry TL, et al. ; Investigators for the Virginia Cardiac Services Quality Initiative . Preoperative anemia versus blood transfusion: which is the culprit for worse outcomes in cardiac surgery? J Thorac Cardiovasc Surg. 2018;156(1):66-74. doi: 10.1016/j.jtcvs.2018.03.109 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Abdullah HR, Sim YE, Sim YT, et al. Preoperative red cell distribution width and 30-day mortality in older patients undergoing non-cardiac surgery: a retrospective cohort observational study. Sci Rep. 2018;8(1):6226. doi: 10.1038/s41598-018-24556-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Aydinli HH, Aytac E, Remzi FH, Bernstein M, Grucela AL. Factors associated with short-term morbidity in patients undergoing colon resection for Crohn’s disease. J Gastrointest Surg. 2018;22(8):1434-1441. doi: 10.1007/s11605-018-3763-4 [DOI] [PubMed] [Google Scholar]
- 6.Sarkiss CA, Hersh EH, Ladner TR, et al. Risk factors for thirty-day morbidity and mortality in extradural lumbar spine tumor resection. World Neurosurg. 2018;114:e1101-e1106. doi: 10.1016/j.wneu.2018.03.155 [DOI] [PubMed] [Google Scholar]
- 7.Siracuse BL, Chamberlain RS. A preoperative scale for determining surgical readmission risk after total hip replacement. JAMA Surg. 2016;151(8):701-709. doi: 10.1001/jamasurg.2016.0020 [DOI] [PubMed] [Google Scholar]
- 8.Nguyen QT, Lee EJ, Huang MG, Park YI, Khullar A, Plodkowski RA. Diagnosis and treatment of patients with thyroid cancer. Am Health Drug Benefits. 2015;8(1):30-40. [PMC free article] [PubMed] [Google Scholar]
- 9.National Cancer Institute Surveillance Epidemiology and End Results Program Cancer stat facts: thyroid cancer. https://seer.cancer.gov/statfacts/html/thyro.html. Accessed May 4, 2018.
- 10.Tanvetyanon T, Choudhury AM. Severity, risk factors, and physician practices in the management of anemia during concurrent chemoradiation for head and neck carcinoma. Cancer. 2006;106(7):1554-1559. doi: 10.1002/cncr.21769 [DOI] [PubMed] [Google Scholar]
- 11.Hollis RH, Singletary BA, McMurtrie JT, et al. Blood transfusion and 30-day mortality in patients with coronary artery disease and anemia following noncardiac surgery. JAMA Surg. 2016;151(2):139-145. doi: 10.1001/jamasurg.2015.3420 [DOI] [PubMed] [Google Scholar]
- 12.Zhu Y, Wang G, Liu S, et al. Risk factors for postoperative delirium in patients undergoing major head and neck cancer surgery: a meta-analysis. Jpn J Clin Oncol. 2017;47(6):505-511. doi: 10.1093/jjco/hyx029 [DOI] [PubMed] [Google Scholar]
- 13.Farwell DG, Reilly DF, Weymuller EA Jr, Greenberg DL, Staiger TO, Futran NA. Predictors of perioperative complications in head and neck patients. Arch Otolaryngol Head Neck Surg. 2002;128(5):505-511. doi: 10.1001/archotol.128.5.505 [DOI] [PubMed] [Google Scholar]
- 14.Papaleontiou M, Hughes DT, Guo C, Banerjee M, Haymart MR. Population-based assessment of complications following surgery for thyroid cancer. J Clin Endocrinol Metab. 2017;102(7):2543-2551. doi: 10.1210/jc.2017-00255 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.American College of Surgeons National Surgical Quality Improvement Program User guide for the 2012 ACS NSQIP participant use data file. https://www.facs.org/~/media/files/quality%20programs/nsqip/ug12.ashx. Published October 2013. Accessed October 13, 2018.
- 16.Kirschner CG. Physicians’ Current Procedural Terminology: CPT. Chicago: American Medical Association; 1996. [Google Scholar]
- 17.American Medical Association. International Classification Of Diseases, 9th Revision, Clinical Modification: Physician ICD-9-CM, 2005: Volumes 1 And 2, Color-Coded, Illustrated, Volume 1. Chicago: American Medical Association; 2004. [Google Scholar]
- 18.Knottnerus A, Tugwell P. STROBE–a checklist to Strengthen the Reporting of Observational Studies in Epidemiology. J Clin Epidemiol. 2008;61(4):323. doi: 10.1016/j.jclinepi.2007.11.006 [DOI] [PubMed] [Google Scholar]
- 19.is Gabriel RA, Clark AI, Nguyen AP, Waterman RS, Schmidt UH. The association of preoperative hematocrit and transfusion with mortality in patients undergoing elective non-cardiac surgery. World J Surg. 2018;42(7):1939-1948. doi: 10.1007/s00268-017-4359-y [DOI] [PubMed] [Google Scholar]
- 20.Desai N, Schofield N, Richards T. Perioperative patient blood management to improve outcomes [published online October 19, 2017]. Anesth Analg. [DOI] [PubMed] [Google Scholar]
- 21.Shander A, Knight K, Thurer R, Adamson J, Spence R. Prevalence and outcomes of anemia in surgery: a systematic review of the literature. Am J Med. 2004;116(suppl 7A):58S-69S. doi: 10.1016/j.amjmed.2003.12.013 [DOI] [PubMed] [Google Scholar]
- 22.Rutledge JW, Spencer H, Moreno MA. Predictors for perioperative outcomes following total laryngectomy: a University HealthSystem Consortium Discharge database study. Otolaryngol Head Neck Surg. 2014;151(1):81-86. doi: 10.1177/0194599814528451 [DOI] [PubMed] [Google Scholar]
- 23.Lebo NL, Caulley L, Alsaffar H, Corsten MJ, Johnson-Obaseki S. Peri-operative factors predisposing to pharyngocutaneous fistula after total laryngectomy: analysis of a large multi-institutional patient cohort. J Otolaryngol Head Neck Surg. 2017;46(1):54. doi: 10.1186/s40463-017-0233-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Michailidou M, Nfonsam VN. Preoperative anemia and outcomes in patients undergoing surgery for inflammatory bowel disease. Am J Surg. 2018;215(1):78-81. doi: 10.1016/j.amjsurg.2017.02.016 [DOI] [PubMed] [Google Scholar]
- 25.Tauriainen T, Koski-Vähälä J, Kinnunen EM, Biancari F. The effect of preoperative anemia on the outcome after coronary surgery. World J Surg. 2017;41(7):1910-1918. doi: 10.1007/s00268-017-3911-0 [DOI] [PubMed] [Google Scholar]
- 26.Mulvey CL, Brant JA, Bur AM, et al. Complications associated with mortality after head and neck surgery. Otolaryngol Head Neck Surg. 2017;156(3):504-510. doi: 10.1177/0194599816686958 [DOI] [PubMed] [Google Scholar]
- 27.von Heymann C, Kaufner L, Sander M, et al. Does the severity of preoperative anemia or blood transfusion have a stronger impact on long-term survival after cardiac surgery? J Thorac Cardiovasc Surg. 2016;152(5):1412-1420. doi: 10.1016/j.jtcvs.2016.06.010 [DOI] [PubMed] [Google Scholar]
- 28.Tohme S, Varley PR, Landsittel DP, Chidi AP, Tsung A. Preoperative anemia and postoperative outcomes after hepatectomy. HPB (Oxford). 2016;18(3):255-261. doi: 10.1016/j.hpb.2015.09.002 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Obi AT, Park YJ, Bove P, et al. The association of perioperative transfusion with 30-day morbidity and mortality in patients undergoing major vascular surgery. J Vasc Surg. 2015;61(4):1000-9.e1. doi: 10.1016/j.jvs.2014.10.106 [DOI] [PubMed] [Google Scholar]
- 30.Phan K, Dunn AE, Kim JS, et al. Impact of preoperative anemia on outcomes in adults undergoing elective posterior cervical fusion. Global Spine J. 2017;7(8):787-793. doi: 10.1177/2192568217705654 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Dai L, Mick SL, McCrae KR, et al. Preoperative anemia in cardiac operation: does hemoglobin tell the whole story? Ann Thorac Surg. 2018;105(1):100-107. doi: 10.1016/j.athoracsur.2017.06.074 [DOI] [PubMed] [Google Scholar]
- 32.Saager L, Turan A, Reynolds LF, Dalton JE, Mascha EJ, Kurz A. The association between preoperative anemia and 30-day mortality and morbidity in noncardiac surgical patients. Anesth Analg. 2013;117(4):909-915. doi: 10.1213/ANE.0b013e31828b347d [DOI] [PubMed] [Google Scholar]