MIAMI BEACH – Cardiac and respiratory comorbidities were "common culprits" and present in more than half of thyroidectomy patients who died in the hospital, according to analysis of a large inpatient database.
Although overall mortality is less than 1% for thyroidectomy patients nationwide, researcher Rishi Vashishta said, "Patient comorbidities can often contribute to perioperative death and should really be considered when discussing treatment options with patients."
Mr. Vashishta and his associates identified 11,862 patients who underwent thyroidectomy using ICD-9 codes from the Healthcare Cost Utilization Project Nationwide Inpatient Sample (NIS) database for 2009. Two-thirds of patients were white and three-fourths were female A total of 73 of these patients died during their hospitalization that year.
"We calculated the mortality rate during hospitalization to be 0.61%," Mr. Vashishta, a medical student at George Washington University, Washington, said at the Triological Society Combined Sections meeting.
Other aims of the study were to assess thyroid surgery complications, length of stay, and total hospital charges. "There are a large number of institutional studies, but there remains a relative paucity of studies examining this procedure on a more macro and socioeconomic level," said Mr. Vashishta.
Among the nearly 12,000 admissions, mean length of stay was 2.97 days and mean total hospital charges accrued was $39,236.
In contrast, a subgroup analysis revealed mean length of stay was 13.8 days and mean increase in total hospital charges was nearly $218,855 among patients who died during hospitalization. "Interestingly, the respiratory status in these patients was markedly worse, with a tracheostomy required in 28%, prolonged mechanical ventilation required in 43%, and endotracheal intubation in 55%," Mr. Vashishta said at the meeting, which was jointly sponsored by the Triological Society and the American College of Surgeons.
Acute cerebrovascular disease was involved in 62% of deaths, he reported.
The mean age of patients who died was 65 years, compared with a mean of 53 years for all thyroidectomy patients in the study.
Approximately 80% of all surgeries in the study were elective. The majority of patients, 55%, underwent total thyroidectomy, 32% underwent unilateral lobectomy, and the remainder had partial thyroidectomy.
When Mr. Vashishta and his colleagues assessed complications, they found hypocalcemia present in 6%, vocal cord paresis in 1.4%, and hypoparathyroidism in 0.77% of patients using bivariate analyses. The incidence of hematoma and hemorrhage were low at 1.43% and 0.67%, respectively. "Our complication rates were generally consistent with those from institutional studies published in the literature."
"We found strong predictors of [these] complications during hospitalization included female gender; hospital location and teaching status; and type of thyroid diagnosis," he said. "Although the majority of cases were conducted at large teaching hospitals in urban centers, no socioeconomic or regional differences were observed," the investigators noted in their abstract but did not offer further explanation.
Admissions data showed that nontoxic nodular goiter was a diagnosis code for 36% of patients. In addition, malignant neoplasm was a code for 31% and benign neoplasm for 11%, "Graves’ disease, which we classified under acquired hypothyroidism, was much less common, around 8%," Mr. Vashishta said. ICD-9 codes for thyrotoxicosis and thyroiditis each were noted on 8% of records.
Errors in coding and sampling are a potential limitation of this and any study based on a large administrative database, Mr. Vashishta said. For example, use of ICD-9 codes "inevitably included patients in our stratified sample admitted for some other problem who underwent incidental thyroidectomies during their hospitalization." Furthermore, thyroidectomy is increasingly being performed as an outpatient procedure and the NIS is an inpatient database. "This effectively skewed our mean total charges and mean length of stay in the hospital upwards."
The study was not funded by industry. Mr. Vashishta said that he had no relevant financial disclosures.