Keywords
Clinical practice guideline - pattern - persistently symptomatic hypothyroidism - primary hypothyroidism - subclinical hypothyroidism - t4/t3 combination - thyroid and pregnancy - thyroid antibodies - thyroxine hypothyroidism - trends
Introduction
Hypothyroidism affects 2%–5% of the general population in several studies from North America and Europe.[[1]],[[2]],[[3]],[[4]] Patients with uncorrected disease suffer significant morbidity and have an increased risk of cardiovascular disease and neurocognitive impairment.[[5]] Patients may seek help and end under the care of various clinical disciplines. Levothyroxine, the treatment of choice, is inexpensive, easy to administer, and in most cases restores well-being while normalizing thyroid function.[[6]] Still, a number of controversies remain, including the treatment threshold in patients with mild or subclinical disease,[[7]] the need for age-specific targets for serum thyroid-stimulating hormone (TSH) levels,[[8]] whether administration of L-T4 alone provides adequate tissue level triiodothyronine (T3) levels,[[9]] optimal control of thyroid function during pregnancy,[[10]] and screening indications for the detection of hypothyroidism in the general population and in women planning pregnancy.[[11]]
Several clinical practice guidelines (CPGs) have recently been published by international organizations dealing with diagnostic evaluation and treatment strategies for adults with hypothyroidism.[[12]],[[13]],[[14]],[[15]],[[16]] Practicing clinicians are usually encouraged to use the recommendations contained in those guidelines in conjunction with their own clinical judgment and within the context of individual patient circumstances.[[12]],[[13]],[[14]],[[15]],[[16]] Attempts have been made to ascertain the extent to which current practices differ from recommendations made in the hypothyroidism CPG.
A couple of studies surveyed the clinical management practices for patients with hypothyroidism in some parts of the world.[[17]],[[18]] Most notable is the international survey by Burch et al.[[17]] This survey included a minority representation (48 responses) from the Middle East and Africa (MEA) region.[[17]] It is anecdotally stated that clinical practice in the MEA region follows international guidelines by virtue of past training and affiliations to North American and European associations. However, this assumption has never been verified. We have therefore conducted this survey to scope the patterns of clinical practice relating to the management of primary hypothyroidism in adults in the MEA region aiming to provide a baseline database to direct future education and research.
Materials and Methods
Objectives
We aimed to (1) document current practices in the general approach to the management of primary hypothyroidism in adults; (2) evaluate management modifications in unique clinical circumstances such as subclinical hypothyroidism, pregnancy, and old age; and (3) compare the practice in MEA with published guidelines.
Target population
In the absence of a single MEA regional endocrine society with a membership list that can define a study population, the target population was identified from a list of electronic mails pooled from continuous professional development delegates, speakers, authors, or members of various scientific groups or forums in various parts of the MEA region. Consequently, a number of questions were added to the survey to help define the demographic professional profiles of the respondents and their practices similar to our previously published.[[19]],[[20]],[[21]],[[22]],[[23]],[[24]],[[25]]
Survey management
A web-based commercial survey management service (SurveyMonkey, Palo Alto, CA, USA) was used. All participants received an initial E-mail that explained the rationale of the survey and what is required from the consented respondents, followed by five subsequent reminder E-mails during the study period. Each message included an explanation of the rationale and method of participation, full credentials, and contact details of the principal investigator together with a unique E-mail-specific electronic link to the questionnaire. Repeated submissions from the same IP address were automatically blocked by the survey service. The survey website was open for the 12-month study period from July 1, 2015, to June 31, 2016. In the end, survey responses were collected and stored electronically for an anonymous analysis.
Survey questionnaire
The questionnaire itself was analogous to the one used in the international survey of 2013 by Burch et al.[[17]] The survey was served in its original English. The survey included questions covering diagnostic evaluation, choice of therapy, and follow-up for an index case of primary overt hypothyroidism followed by three clinical variants [[Table 1]]. The variants include (1) a patient with persistent symptoms of hypothyroidism despite having achieved target thyroid hormone levels, (2) a patient anticipating pregnancy, and (3) a patient with subclinical hypothyroidism [[Table 1]].
Table 1: The clinical details of the index case and the variants
Data analysis
Summary statistics were prepared for responses to each question. Because not every participant answered all questions, the percentage of respondents providing a given answer was calculated individually for each question, using the number of respondents to that question in the denominator.
Results
Participants' demographic and professional profiles
Four groups represented the majority of respondents, namely 113 adult endocrinologists (36.1%), 54 physicians in general internal medicine (17.3%), 51 primary care physicians (16.3%), and 49 general internal physicians with special interest and practice in endocrinology (15.7%). The remainder were pediatric endocrinologists (9.0%), obstetrician/gynecologist (3.2%), surgeons (1.9%), and nuclear medicine physicians (0.64%). The majority (82.2%) were in clinical practice for over 10 years (34.9% for 10–20 years and 47.8% for over 20 years) and 17.3% have been practicing for <10 years. Over half (55.0%) of respondents hold senior consultant/attending physician positions, 27.2% were middle-grade physicians, and the remainder were in other grades. Most respondents (77.4%) were practicing clinicians working mainly in clinical health services and 22.6% have full-time university appointments. Reported professional affiliations included membership of the American Association of Clinical Endocrinologists and local chapters (50.6%), regional endocrine societies (45.8%), the US Endocrine Society (22.0%), European Society of Endocrinology (16.1%), American Thyroid Association (10.7%), UK Society for Endocrinology (6.0%), and European Thyroid Association (3.0%). Access to thyroid-related diagnostic and treatment resources as indicated clinically was reported as follows: measurements of thyroid hormones (97.4%), thyroid ultrasound imaging (90.5%), thyroid antibodies (86.2%), antithyroid drug therapy (84.6%), fine-needle aspiration and cytological examination (68.2%), thyroid surgeon (65.3%), an ophthalmologist with expertise in thyroid eye disease (55.4%), and nuclear medicine scanning (49.5%).
Diagnostic evaluation of the index case
[[Figure 1]] shows the percentage of respondents ordering the listed laboratory tests for the index case compared to the 2013 international survey.[[17]] The most remarkable difference was a higher use of ultrasound examination in hypothyroid patients undergoing initial evaluation (66.9% MEA versus 44.4% in the whole of the 2013 international survey).[[17]] Greater request for measurement of lipid profiles was also noted followed by measurements of antithyroglobulin antibodies and serum T3. Repeating measurements of TSH and T4 and measuring thyroid peroxidase (TPO) antibody (Ab) testing was not remarkably different in MEA from the rest of the world [[Figure 1]].
Figure 1: Percentage of participants who would request the listed testing in a patient with overt hypothyroidism. The Middle East and Africa data (closed) are from the present study and the international 2013 survey data (open) are from reference 17
Treatment choices for the index case
Thyroid hormone preparation
Out of 355 respondents, 71.6% would use a brand name of L-T4, and 25.6% would use a generic formulation for the initial therapy of overt hypothyroidism [[Figure 2]]. Six respondents (1.7%) selected combined L-T4 and L-T3 as initial therapy in the index case. Only three respondents (0.9%) selected thyroid extract as initial therapy. [[Table 2]], [[Figure 1]] and [[Figure 2]] shows the responses in the MEA regions and the international survey of 2013. There was greater use of brand-name formulations in MEA (71.6%) compared with Latin America (58.0%), Europe (58.8%), Asia-Oceania (58.2%), and North America (37.9%) [[Figure 2]].
Figure 2: The percentage responses reflect choices of thyroid hormones for initial therapy. The Middle East and Africa data (closed) are from the present study and the international 2013 survey data (open) are from reference 17
Table 2: The concordance of survey responses with select subset of the 2012 clinical practice guidelines recommendations in the international survey (American Thyroid Association/ES/American Association of Clinical Endocrinologists 2013)a (American Thyroid Association 2013) and the present study (Middle East and Africa 2015)
Follow-up
Among 356 respondents, 34.6% would recheck thyroid hormone levels 6 weeks after starting thyroid hormone therapy, followed by 4 weeks (29.2%), 8 weeks (18.5%), 2 weeks (9.8%), and 12 weeks (5.9%). Specific testing at the time of follow-up included TSH for 340 of 348 (97.7%) respondents, free T4 (53.5%), free T3 (18.1%), or total T3 (4.9%).
Thyroid-stimulating hormone targets
Among respondents, the target TSH in the 52-year-old index case was 2.0–2.9 mU/L in 118 responses (34.5%), 1.0–1.9 mU/L for 55 respondents (15.9%), 3.0–3.9 mU/L in 90 respondents (26%), and 0.5–0.9 mU/L in 55 respondents (5.2%). When asked what the TSH target would be in a 25-year-old patient, there was a shift toward lower TSH values, with 38.3% now selecting 2.0–2.9 mU/L, followed by 1.0–1.9 mU/L (25.5%), 3.0–3.9 mU/L (14.9%), 0.5–0.9 mU/L (9.7%), and 4.0–4.9 mU/L (7.0%). Conversely, in an 85-year-old patient, 33.5% would select a TSH from 4.0–4.9 mU/L, followed by 3.0–3.9 mU/L by (25.0%), 2.0–2.9 mU/L, and 5.0–9.9 mU/L (12.8% each). A comparison of TSH targets for the three age variants is shown in [[Figure 3]].
Figure 3: Age-specific thyroid-stimulating hormone targets among survey participants (upper) for 25 years old, 52 years old, and 85 years old
Therapy for overt hypothyroidism
Among 366 respondents, 350 (97.6%) would initiate thyroid hormone therapy in the index case of overt hypothyroidism compared to 98.9% in the international survey. We queried respondents about their usual technique for correcting overt hypothyroidism. Among 367 respondents to this question, 244 (66.5%) would gradually restore euthyroidism, 54 (14.7%) would select an empiric dose adjusted to achieve target levels, and 68 (18.5%) would start with a calculated full-replacement dose. A greater proportion of the Middle East-Africa (66.5%) respondents would use a gradual approach compared with Latin America (60.5%), Europe (55.8%), Asia-Oceania (55.1%), and North America (30.5%).[[13]] For respondents preferring a gradual restoration of euthyroidism, most (56.6%) would increase in increments of 25 μg, followed by 50 μg (21.7%) and 12.5 μg (17.8%). The frequency of incremental increases was 6 weeks (34.6%) followed by 4 and 8 weeks by 29.2% and 18.2%, respectively. Long-term follow-up: After achieving stable target TSH values, respondents were asked how often they would repeat thyroid laboratory testing. Among 348 respondents, 47.4% would obtain laboratory studies at 6-month intervals, followed by 3 months (35.3%), 12 months (11.8%), and <3 months (5.2%). The manner in which asymptomatic patients at target TSH values would be followed was by laboratory studies plus office visits by 59.7% of 348 respondents, return to the primary care physician by 33.1%, or laboratory studies plus a phone call by 7.0% of respondents.
Variation 1: Persistent hypothyroid symptoms
When a patient despite achieving target TSH values on L-T4 therapy still has persistent hypothyroid symptoms, was presented to participants [[Table 1]], of 329 respondents, 86.9% would perform testing for other sources of the patient's symptoms, 5.8% would add L-T3 therapy to L-T4, 4.6% would increase the dose of the patient's L-T4 and 1.5% would refer the patient to behavioral health but only 1.2% would refer the patient back to their primary care physician for further evaluation. Additional testing requested in patients with persistent unexplained hypothyroid symptoms would include a complete cell count by 86.0% of 286 respondents, a complete metabolic panel (71%), morning cortisol level (67.5%), serum Vitamin B 12 levels (57.7%), or T3 levels (25.5%). In particular, measurement of 25-hydroxyvitamin D levels was suggested by 13 respondents only.
Variation 2: Hypothyroidism and pregnancy
Respondents were queried regarding their approach to a 25-year-old woman with Hashimoto's thyroiditis, who is planning pregnancy and has a serum TSH of 3.5 mU/L on L-T4 therapy [[Table 1]]. For the prepregnant patient, the preferred TSH target for 316 respondents was 2.0–2.4 mU/L (28.5%), followed by 1.5–1.9 mU/L (15.8%), 0.5–0.9 mU/L (13.6%), 1.0–1.4 mU/L (13.0%), and 2.5–3.0 mU/L (7.9%). Respondents were next asked about immediate changes in L-T4 dose after a confirmed pregnancy in a woman with a recent TSH value of 0.5 mU/L [[Figure 4]]. Among 316 respondents, 52.2% would continue the current dose of L-T4, 23.7% would increase the dose by one-third, 14.9% would decrease the dose by one-third, 4.8% would increase the dose by 50%, and 4.4% would decrease the dose by 50%. The preferred TSH range during pregnancy was similar to the prepregnant patient, with a TSH target of 2.0–2.4 mU/L among 25.4% of 315 respondents, 0.5–0.9 mU/L (16.2%), 1.5–1.9 mU/L (16.5%), 1.0–1.4 mU/L (11.8%), and 0.1–0.4 mU/L (3.2%). However, an unexpectedly large proportion (27.0%) preferred TSH >2.5 mU/L [[Figure 4]]. Among 318 respondents, 62.9% would check thyroid laboratory studies every 4 weeks during pregnancy, 17.6% every 8 weeks, 10.7% every 12 weeks, and 8.8% every 2 weeks.
Figure 4: The widely distributed range of thyroid-stimulating hormone targets for pregnant and prepregnant patients. An unacceptable long tail of responses choosing target thyroid-stimulating hormone >2.5 mIU/L
Variation 3: Subclinical hypothyroidism
A case of subclinical hypothyroidism in an asymptomatic 52-year-old woman with a serum TSH of 7.7 mU/L was presented to participants [[Table 1]]. Among 306 respondents, 82.0% would perform TPO Ab testing, 60.8% would request a thyroid ultrasound, 58.2% would obtain a lipid panel, 43.8% would obtain antithyroglobulin Ab, and would order either a free T3 (38.9%) or total T3 levels (11.8%). Among 300 respondents, the percentage that would start L-T4 therapy for subclinical hypothyroidism in the presence of the listed clinical factors is shown in [[Figure 5]]. Most respondents indicated more than one potential indication for treatment. Treatment without additional justification was selected by 9.0% of respondents from the Middle East-Africa. This was lowest among previously reported response rates from Asia-Oceania (9.3%) and Latin America (10.8%) but less than responses from Europe (14.3%) and North America (24.7%).[[17]]
Figure 5: Percentage of participants who would initiate treatment of subclinical hypothyroidism in the presence of the listed circumstance in the Middle East and Africa regions (closed) and the international 2013 survey (open)
Discussion
The current report provides results of a survey of clinical practices in the management of primary hypothyroidism by 368 physicians practicing in the MEA region. A previous survey of the international members of American endocrine societies included a minority representation (48 responses) from this region.[[17]] We have used an analogous survey instrument used in the international survey to allow making comparisons of the present data set with the previously published survey.[[17]] This is part of a project aiming to document patterns of clinical practice in the management of common endocrine conditions in the MEA.[[20]],[[21]],[[22]],[[23]],[[24]],[[25]],[[26]]
The salient findings are first nearly exclusive preference for L-T4 alone as initial therapy, in line with current practice, and second, a low threshold for treating mild thyroid failure in line with European and American recommendations. There is excessive use of US scanning in the MEA region (66.9%) compared to the international cohort (44.4%); it is likely unnecessary in most cases without a palpable goiter. This finding is similar to a recent survey from India.[[19]]
More respondents from the MEA region opted that there is a highly variable approach to both the rate and means of restoring euthyroid status for overt disease. The careful cautious approach may be due to the influence of older teaching or a reflection on the patient population encountered in the practice with more advanced disease and comorbidities.[[27]] Although 25 mcg tablets are available, clinicians have learned to escalate the dose in as low as 12.5 mcg doses because of the very long half-life of T4. The frequency of incremental increase of 6–8 weeks by more than 50% of respondents is in line with current guidelines. Long-term follow-up reflects practice appropriate for the region and 6 months is the usual interval. A combination of T3 and T4 is infrequently used. This could reflect the controversy surrounding this issue, level of knowledge or preference of convenience, or inconsistent access to T3 in general.[[28]] However, the very low use of thyroid extracts most likely reflects nonavailability or general avoidance of porcine products (particularly when consumed orally) in Muslim-majority countries although the differences are not particularly remarkable between the regional and international data in this respect[[17]] [[Figure 1]]b.
In patients with persistent hypothyroid symptoms, most respondents would perform testing for other possible causes of patients' symptoms and only a small proportion would add L-T3 therapy to LT-4, consistent with the fact that only recently has the medical community agreed that may be helpful in a small proportion of such patients.[[28]]
There is some inconsistent tendency toward age-specific TSH targets, and L-T4 adjustments made using 25 μg increments, to achieve a target TSH value of 1–2 mU/L in younger patients and 2–5 mU/L in an octogenarian.[[2]],[[29]]
Management of established hypothyroidism in the pregnant or prepregnant woman has been addressed in recent CPGs.[[30]],[[31]] There is general agreement that L-T4 requirements increase in most pregnant women, often by up to 50% above baseline requirements despite some variations.[[32]],[[33]] Furthermore, due to concerns regarding adverse neurodevelopmental consequences associated with maternal hypothyroidism during early pregnancy,[[34]],[[35]],[[36]] lower TSH targets are recommended in this setting.
The TSH targets in pregnant and prepregnant women recommended by respondents are widely variable and truly complacent [[Figure 4]], suggesting a lack of knowledge of nonobstetric physicians of the guidelines or no real actual involvement in the care of these women. The TSH targets for the first trimester are usually low, and the increase in dose is in anticipation of increased deiodinase activity and increased metabolic rate and cardiac output; this was not obvious from our data [[Figure 4]]. Despite the current recommendation to increase the dose of L-T4 by 2 pills per week at pregnancy confirmation in the hypothyroid patient,[[30]],[[31]],[[32]],[[33]] our patient with a TSH value of 0.5 mU/L at pregnancy diagnosis would be maintained on the same dose by more than half of respondents (52.2%).
Patients with subclinical hypothyroidism represent a subset in which treatment with thyroid hormone is controversial.[[7]],[[37]] Nonetheless, 9% of respondents would treat such a patient with thyroid hormone without further justification, although many more would like to have justifications to initiate treatment [[Figure 5]]. Overtreatment with thyroid hormone may result in indiscriminate treatment with thyroid hormones. It is a frequent finding in patients intended for replacement therapy. The resultant subclinical thyrotoxicosis has been linked to adverse consequences such as atrial fibrillation and osteoporosis.[[38]] The survey questionnaire could not deduce the reasons behind treatment.
Our study has both strengths and limitations. The number of respondents from the MEA region is much higher than those included in the international survey[[17]] making it more representative of the region. The electronic mail invitation to participate provided an opportunity for more than 4000 potential respondents to learn about the survey and respond honestly at their own time. However, limitations of the present study are convenience sampling, loose definition of the recipient population, and selection bias. Another limitation is the inhomogeneous representation of various countries with marked diversity in the availability of treatment options. Selection bias could result in respondents to the survey being more aware of CPGs and potentially more adherent to their recommendations, thus overestimating the quality of care provided. Finally, the international differences noted in management practices may be a reflection of resource availability rather than the knowledge and skills of the physicians.
Conclusions
The present survey of physicians on the management of primary hypothyroidism documented for the first time current practice patterns and demonstrates both alignment and focal deviation from current CPGs. It should help in directing focused educational activities in the region, providing a baseline for future monitoring of practices. Regular revisiting using the same, modified, or totally updated instrument is warranted.
Authors contribution
SAB and IHS proposed the idea of the study. SAB adapted the survey and drafted the manuscript. All authors examined the results and reviewed the manuscript for intellectual content and style. They all approved the final manuscript.
Compliance with ethical principles
The study was approved by the Institutional Review Board of Sheikh Khalifa Medical City, Abu Dhabi, UAE. All participants provided electronic informed consent to participate before they could proceed to access the survey questions. All data were analyzed anonymously.
Data availability
Data will be available by reasonable requests to the corresponding author.
