This study calculated post-thyroidectomy calcium treatment rates to investigate a correlation between the season of surgery and post-thyroidectomy hypocalcemia. The overall rate of patients receiving post-operative calcium treatment before discharge, in the study population, was 32.6%. Given the wide range of rates of post-operative hypocalcemia reported in the literature, it is reasonable to say that this is in line with previously reported results [6].
It was found that post-operative calcium treatment was given significantly less frequently to patients operated at the end of summer and beginning of fall than patients operated at the end of winter and beginning of spring. The results obtained in this study supports findings previously presented by Mascarella et al. [18]. However, they used symptoms of hypocalcemia combined with plasma concentration of calcium instead of calcium treatment as their primary variable. On the other hand, the present study design made it possible to include a considerably larger sample size. Our studies also differed slightly in how the patients were stratified according to season.
In the next step, we aimed to investigate whether the previously discussed seasonal difference was persistent over time. Since neither calcium levels nor parathyroid hormone (PTH) are routinely measured on all patients after thyroidectomy at Swedish hospitals, prescription of vitamin D due to hypoparathyroidism and calcium supplements due to hypocalcemia, at discharge and at six-month follow-up were used as surrogate variables for persistent hypocalcemia. Calcium supplements were, in fact, prescribed more frequently to patients in the dark group at 6 months, as compared to the bright group. These rates are slightly higher compared to previously reported rates of persistent hypocalcemia [6]. The correlation was the same for vitamin D supplements; however, the difference was not statistically significant. This indicates that the observed seasonal difference in post-operative calcium treatment may also be true for persistent parathyroid failure. Since calcium supplement alone is rarely enough to correct plasma calcium levels when PTH is absent due to hypoparathyroidism, the patients need treatment with active Vitamin D. At follow-up, the surgeons are considering the registry wording “Vitamin D treatment due to hypoparathyroidism” which in the vast majority of cases means treatment with active Vitamin D. Moreover, since the treatment for hypocalcemia was not only temporary but prolonged at follow-up, this could be of importance in the clinical setting, since the patients then would be in need not only for surveillance of the levothyroxine substitutions but also of calcium medication, with possible impact on quality of life. Furthermore, data on prescribed supplements at the six-months follow-up were missing in 11% of the patients in the dark group and in 14% of the patients in the bright group, which may affect the result to some extent. However, we argue that this is a non-differential misclassification, and an imputation analysis providing the missing values with estimates, verified the results.
This retrospective register-based study does not provide any evident cause of the seasonal variation observed. Even though several factors could influence the correlation, we find it plausible that the naturally occurring vitamin D levels play an essential role. The present study was conducted on a large, nation-wide, study population. This large cohort made it possible to stratify on both season and indication for surgery with a sample size large enough in every subgroup. The study population was divided into groups based on the expected highest and lowest vitamin D levels. To verify that the levels of vitamin D are, in fact, responsible for the differences between the dark and the bright group, it would be necessary to correlate actual vitamin D values to the risk for post-operative hypocalcemia. This has previously been done, but not to our knowledge correlated to season [4, 10].
The seasonal difference in the rate of post-operative calcium treatment was still seen after stratifying on indications for surgery, both goiter and thyrotoxicosis. Interestingly, no seasonal difference was found if the indication for surgery was confirmed malignancy or suspicion of malignancy. One can speculate that the selection of cancer patients may influence this to specific units and/or surgeons. Furthermore, the difference in the biological nature of the various indications for surgery may affect calcium metabolism. In a previously published prospective observational study “hungry bone syndrome” caused by mineral depletion of bone due to increased bone metabolism during the period of thyrotoxicosis was found to be the major cause of hypocalcemia after total thyroidectomy [9]. The fact that no seasonal variation was seen if the indication for surgery was malignancy may be influenced by the unequal distribution of age, since high age is associated with a higher risk of post-operative hypocalcemia [12]. Although not statistically significant, mean age was slightly higher in the bright group which could potentially mask the seasonal difference seen in the subgroups.
One can argue that the seasonal variation in calcium treatment may be influenced by different levels of surgical skill over the year or that more difficult cases are prioritized during the summer. However, in Sweden thyroid surgery is only performed by specialized ENT- and endocrine surgeons irrespective of the time of the year, and medical staffing (physicians, anesthesiologists and operating assistants) is largely the same throughout the year, as a structural bias as a result of this can probably not explain seasonal differences. Furthermore, a predominance of more difficult cases during the summer could lead to higher rates of post-operative hypocalcemia and, if anything, diminish the observed seasonal variation.
As mentioned above, actual Ca2+ values were not used to measure hypocalcemia, which may have been intuitive. This was because analysis methods and cut-of values may differ over time and between laboratories; this definition was not considered suitable for this study set up. It was also noted that Ca2+ concentrations were not documented for a large proportion of the study population. Consequently, it was decided to use post-operative calcium treatment as the primary variable. A potential advantage of this definition is that it might include patients with symptoms of hypocalcemia but with Ca2+ in the lower normal range of reference. Of course, this makes it possible to let subjective decisions and the surgeon’s preference influence the results. However, national guidelines on handling post-operative hypocalcemia has been developed to reduce the risk of patients being treated differently depending on operating units and individual surgeons. Furthermore, the large sample size in this study could probably compensate for differences of this nature.