Selective serotonin reuptake inhibitors (SSRI) have been claimed to negatively affect the thyroid function, albeit the evidence is controversial. We searched for studies that measured parameters of thyroid function (TSH, T4, Free T4, or T3) before and after a course of SSRI treatment in euthyroid patients with major depressive disorder. Electronic searches were conducted on MEDLINE, Embase and Web of Science databases from inception through April 4th, 2018. We performed random-effects meta-analyses to estimate the effect of SSRIs on each hormone. A total 1791 records were identified in the electronic search, and 14 observational clinical studies were included in the analyses. All studies had at least moderate risk of bias and were considered of low quality. A course of SSRI treatment was associated with a decrease in T4 of -6.58 nmol/L (95% Confidence Interval [CI], -12.17 to -.99, p = .005, I2=97%; Cohen's d = .50), a decrease in Free T4 of -.91 pmol/L (95% CI, -1.65 to -.16, p = .017, I2=96%; Cohen's d = .66), and a decrease in T3 of -.10 nmol/L (95% CI, -.18 to -.03, p = .007, I2=96%; Cohen's d = .45), and no effect on TSH (0.06 microIU/L, 95% CI, -.05 to .17, p = .285, I2=98%; Cohen's d = .17). We did not detect publication bias in any of the four meta-analyses. We conclude that there is preliminary evidence that SSRIs slightly decrease thyroid function, but quality of evidence is low. Clinical magnitude of such effect is yet unclear.

中文翻译：

---

选择性5-羟色胺再摄取抑制剂对重度抑郁症患者甲状腺功能的影响：系统评价和荟萃分析

据称选择性血清素再摄取抑制剂 (SSRI) 会对甲状腺功能产生负面影响，尽管证据存在争议。我们检索了测量甲状腺功能参数（TSH、T4、游离 T4 或 T3）在患有重度抑郁症的甲状腺功能正常患者接受 SSRI 治疗前后的研究。从开始到 2018 年 4 月 4 日，在 MEDLINE、Embase 和 Web of Science 数据库中进行了电子搜索。我们进行了随机效应荟萃分析，以估计 SSRIs 对每种激素的影响。在电子搜索中总共确定了 1791 条记录，分析中包括了 14 项观察性临床研究。所有研究至少有中等偏倚风险，被认为是低质量的。一个疗程的 SSRI 治疗与 T4 下降 -6 相关。58 nmol/L（95% 置信区间 [CI]，-12.17 至 -.99，p = .005，I2=97%；Cohen's d = .50），游离 T4 降低 -.91 pmol/L（ 95% CI，-1.65 至 -.16，p = .017，I2=96%；Cohen's d = .66），T3 降低 -.10 nmol/L（95% CI，-.18 至 - .03，p = .007，I2=96%；Cohen's d = .45），对 TSH 没有影响（0.06 microIU/L，95% CI，-.05 到 .17，p = .285，I2=98 %；科恩的 d = .17)。我们在四项荟萃分析中的任何一项中均未发现发表偏倚。我们得出结论，初步证据表明 SSRIs 会轻微降低甲状腺功能，但证据质量低。这种影响的临床程度尚不清楚。-.18 至 -.03，p = .007，I2=96%；Cohen's d = .45)，并且对 TSH 没有影响（0.06 microIU/L，95% CI，-.05 到 .17，p = .285，I2=98%；Cohen's d = .17）。我们在四项荟萃分析中的任何一项中均未发现发表偏倚。我们得出结论，初步证据表明 SSRIs 会轻微降低甲状腺功能，但证据质量低。这种作用的临床程度尚不清楚。-.18 至 -.03，p = .007，I2=96%；Cohen's d = .45)，并且对 TSH 没有影响（0.06 microIU/L，95% CI，-.05 到 .17，p = .285，I2=98%；Cohen's d = .17）。我们在四项荟萃分析中的任何一项中均未发现发表偏倚。我们得出结论，初步证据表明 SSRIs 会轻微降低甲状腺功能，但证据质量低。这种作用的临床程度尚不清楚。