Increased Morbidity in Males Diagnosed With Gynecomastia: A Nationwide Register-based Cohort Study

Abstract

Context

Evidence on the long-term and general health of males with gynecomastia is lacking.

Objectives

To assess health before and following a diagnosis of gynecomastia.

Methods

A register-based cohort study of 140 574 males, of which 23 429 were diagnosed with incident gynecomastia and age- and calendar-matched (1:5) to 117 145 males without gynecomastia from the background population. Males with gynecomastia were stratified into males without (idiopathic) or with a known preexisting risk factor (disease/medication). Cox and logistic regression models investigated associations of disease risk according to International Classification of Diseases 10th revision sections following and before gynecomastia diagnosis.

Results

A total of 16 253 (69.4%) males in the cohort were identified with idiopathic gynecomastia. These males had a statistically significant higher risk of future disease across all included disease chapters (hazard ratio [HR], 1.19-1.89), with endocrine diseases representing the greatest disease risk (HR, 1.89; 95% CI, 1.76-2.03). The highest subchapter disease risk was observed for disorders of the endocrine glands (odds ratio [OR], 7.27; 95% CI, 6.19-8.54). Similarly, the ORs of comorbidities were higher across all included disease sections (OR, 1.05-1.51), except for psychiatric disease (OR, 0.72; 95% CI, 0.68-0.78), with the highest association with musculoskeletal/connective tissue (OR, 1.51; 95% CI, 1.46-1.57) and circulatory (OR, 1.36; 95% CI, 1.29-1.43) diseases.

Conclusions

The presence of idiopathic gynecomastia is an important first clinical symptom of an underlying disease and a significant predictor of future disease risk. These findings should stimulate more awareness among health care providers to increase identification of gynecomastia and its causes in males.

Gynecomastia, the benign enlargement of male glandular breast volume, is a common disorder that affects 32% to 65% of the male population depending on age and applied diagnostic criteria (1, 2). Gynecomastia is derived from the Greek terms “gynaíka” (meaning woman) and “mastós” (meaning breast) and most often occurs during periods with changes in levels or activity of sex steroids (1), resulting from increased estradiol (E2) stimulatory activity at estrogen receptors in the male breast (3). True gynecomastia with the presence of glandular breast tissue is distinguished from pseudo-gynecomastia (also termed lipomastia) that is caused by an excessive amount of breast adipose tissue (2).

Gynecomastia has 3 distinct age peaks during a life course: infancy, puberty, and older age. In infancy, an estimated 70% of newborn boys experience transient gynecomastia caused by placental transfer of maternal estrogens or high testosterone secretion in the early postnatal activation of the hypothalamic-pituitary-testis hormone axis in minipuberty (4, 5), which is subsequently aromatized to E2 (6). With the onset of puberty, about 50% to 70% of boys show transient gynecomastia, of which 20% present with an intermittent form (appearing, disappearing, and reappearing) (7). Gynecomastia is most frequently diagnosed in middle-to-older aged males (>40 years), typically related to reduced androgen production, and often attributed to increased adiposity with aging (8, 9). During infancy and puberty, gynecomastia usually resolves spontaneously within weeks to years, whereas in older males, it may be persistent (1).

The incidences of gynecomastia and associated health risks have not been fully elucidated (1), but we recently discovered an increased incidence of gynecomastia in Denmark over the past 20 years (10). The reasons for this trend remain speculative (11), but increasing obesity is likely to play a role in these figures, and some cases may be pseudo-gynecomastia misclassified as gynecomastia. Of note, the International Classification of Diseases, 10th revision (ICD-10), codes used in most countries for gynecomastia are rather unspecific and include breast hypertrophy from any cause, thereby increasing the possibility of bias from misclassification. Importantly, the prospective impact of gynecomastia on long-term health has received very limited attention. Some published data have proposed various mental health and psychological consequences of gynecomastia (12), but no previous studies have investigated the prospective disease risk following gynecomastia nor characterized comorbidities of males at the time of gynecomastia diagnosis.

With unique access to nationwide register-based information on all diseases in males with gynecomastia, we aimed to assess previous morbidities and future disease risks associated with gynecomastia and whether future disease risks were conditional on recognized underlying pathologies. We applied ICD-10 code ABC extensions available in Denmark that allow for a specific diagnosis of gynecomastia (N62.9A), limiting the likely misclassification of gynecomastia occurring in studies from other countries.

Methods

Study Design and Data Source

A cohort study using Danish nationwide health registries was undertaken to assess past and future diseases associated with gynecomastia. The unique 10-digit personal identification number registered in the Danish Civil Registration System allows accurate data linkage between registries (13). We used data from the Danish National Patient Registry that contains records of all individual-level patient discharges from private and public hospitals and treatments in hospital-based outpatient clinics (since 1995) (14). In Denmark, the World Health Organization's ICD-10 has been used since January 1, 1994 (previously ICD-8) (15). Data were also linked to the Danish Cancer Registry (ICD-10 since January 1, 1994, and previously ICD-7) (16) and to the Danish National Prescription Registry (17, 18). Reporting to these registries is compulsory.

The Cohort (n = 140 574)

We created a cohort based on all males diagnosed with gynecomastia from January 1, 1995, to June 30, 2021 (n = 23 429). Each of these males were age- and calendar-matched to 5 randomly selected gynecomastia-free males from the background Danish population (n = 117 145), all alive and living in Denmark at the time of matching. The final cohort included 140 574 males.

Gynecomastia Ascertainment

Exact data of gynecomastia diagnosis were obtained from the Danish National Patient Registry that follows current international standards for diseases classification. In Denmark, the ICD-10 version sometimes has ABC extensions added to specific diagnostic codes, making the Danish version of the ICD-10 more detailed than the international ICD-10 (19). The specific identification of an incident gynecomastia diagnosis (primary or secondary) was based on the Danish ICD-10 code with ABC extensions added: N62.9A (19) (emergency, in- or outpatient hospital contact), which excludes other forms of breast enlargements (hypertrophy of breast, massive pubertal hypertrophy of breast) otherwise included in the ICD-10 N62. This available ABC extension with more specific diagnosis codes limits the risk of diagnosis errors and misclassification of gynecomastia.

The males were stratified into 2 groups: (1) idiopathic (unknown cause) or (2) gynecomastia with a known preexisting condition or use of medication recognized to be associated with the risk of gynecomastia (hereafter, we use the term “gynecomastia with a known risk factor” throughout the manuscript and tables). A male was identified with gynecomastia with a known risk factor (group 2) if there was at least 1 underlying pathologic diagnosis (eg, testosterone deficiency, Klinefelter syndrome, liver insufficiency, kidney insufficiency, malignant neoplasm, alcohol addition, obesity, cannabis addiction) before/on the date of gynecomastia diagnosis, or if a first prescription had been retrieved for specific medications that can cause gynecomastia (antiandrogens, testosterone-5-α reductase inhibitors, aldosterone antagonists, hormones). The identification of risk factors was based on existing knowledge and anecdotal reports (1, 20). See specific diagnoses and medications in Supplementary Table S1 (21).

Morbidities

Detailed information on all morbidities and cancers were obtained from the Danish National Patient Registry and Danish Cancer Registry for the entire cohort. All incident disease admissions (primary or secondary diagnosis)/records of cancers before June 30, 2021 were identified in each of the following 13 ICD-10 chapters: infectious (A00-A99), cancers (C00-D48), blood (D50-D64), endocrine (E00-E90), psychiatric (F00-F99), neurological (G00-H59), circulatory (I00-I99), pulmonary (J00-J99), gastrointestinal (K00-K93), skin (L00-L99), musculoskeletal/connective tissue (M00-M99), diseases in newborns (P00-P96), and congenital diseases (Q00-Q99). The ICD section on urogenital (N00-N99) covered identification of gynecomastia and was not included to identify additional morbidities. To reduce multiple testing and risk of chance findings, we a priori excluded 5 ICD-10 chapters because of lack of relevance (specified in Supplementary Table S2 (21)). In addition, we post hoc excluded the ICD-10 section of diseases in newborns (P00-P96) because of a very limited number of events (n < 20), leaving 12 ICD-10 chapters for statistical analyses.

Statistical Analyses

The crude distribution of incident gynecomastia, stratified by idiopathic or overall and according to predefined age groups (<10, 10-18, 19-40, >40 years), was reported to provide descriptive statistics of the cohort. Age groups were based on expected peak incidence trends of gynecomastia.

In our main analyses, Cox proportional hazards regression models investigated disease risks following gynecomastia diagnosis (disease ICD chapter group-specific incidence) as a function of gynecomastia (idiopathic, overall). We stratified the analyses using each gynecomastia subject and the matched referent subjects (no gynecomastia) as a stratum to ensure that comparisons were adjusted for age and calendar time. For each specific disease ICD chapter group, males with a record of that particular disease before gynecomastia diagnosis were excluded from the analysis. Chapter-specific hazard ratios (HRs) were calculated. The follow-up period began on the date of gynecomastia and ended for each disease strata on the date of incident disease (event), death, emigration/disappearance, or end of follow-up on June 30, 2021, whichever came first. We repeated the main analyses as a function of gynecomastia with a preexisting known risk factor.

To visualize the estimated disease risks, all HRs for gynecomastia (idiopathic, overall) were graphically presented according to diseases chapters, both overall and investigating effects as a function of age (according to 10-year age groups) at the time of the gynecomastia diagnosis.

In secondary analyses, morbidities before gynecomastia diagnosis were assessed by applying logistic regression models using gynecomastia (idiopathic, overall) as the dependent variable and incident disease before/or on the same date as gynecomastia (disease group approach) as the independent variable. Chapter-specific odds ratios (ORs) were calculated. We also repeated these analyses as a function of gynecomastia with a known risk factor. All supplementary analyses (Supplementary Tables S3, S5, and S6) are available in the Dryad Digital Repository (21).

All data management and analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA) and R 3.6.0 (R Foundation for Statistical Computing, Vienna, Austria). Specifically, PROC PHREG and LOGISTIC in SAS and the plotrix package in R were used for analyses and figures, respectively.

Ethical Considerations

The study was approved by the Danish Data Protection Agency (J.nr. P-2020-525). According to Danish law, projects based solely on registry data are exempt from ethical approval by an ethics review board (22). The study was reported according to the Strengthening the Reporting of Observational Studies guidelines and checklist.

Results

Descriptive Characteristics

Age frequencies in the cohort according to gynecomastia are presented in Table 1. In total, 23 429 (16.7%) males in the cohort were diagnosed with gynecomastia, of which 16 253 (69.4%) were identified with idiopathic gynecomastia and predominantly aged 19 to 40 years (44.3%). The distribution of known risk factors to gynecomastia within age group strata is presented in Supplementary Table S3 (21). The predominant existing risk factors (all ages) were medication (52.3%), alcohol addiction (15.7%), testosterone deficiency (9.7%), malignant neoplasm (10.0%), and obesity not specified with endocrine origin (9.6%).

Table 1.

Age frequencies in the cohort (n = 140 574) according to gynecomastia (idiopathic^a, overall)

	Gynecomastia
	Yes (23 429)				No (117 145)
	Idiopathica (n = 16 253)		Overall (n = 23 429)		References
Age, yb	n	%	n	%	n	%
<10	84	0.52	92	0.39	459	0.39
10-18	2525	15.54	2775	11.84	13 901	11.87
19-40	7194	44.26	8720	37.22	43 577	37.20
>40	6450	39.68	11 842	50.54	59 208	50.54

This group of males did not have a known risk factor before the gynecomastia diagnosis.

Age ascertained at the time of baseline (index date for gynecomastia and reference population).

Disease Risks After Gynecomastia Diagnosis

Disease risks during follow-up after a diagnosis of gynecomastia are presented in Table 2 (by ICD chapters), Supplementary Table S4 (by ICD subtypes) (21) and visualized in Fig. 1. In general, males with idiopathic gynecomastia had a statistically significant higher risk of disease across all included chapters (HR, 1.19-1.89). The highest incident disease risk for males after a diagnosis of idiopathic gynecomastia was endocrine diseases (HR, 1.89; 95% CI, 1.76-2.03). This was mainly driven by disorders of endocrine glands (E20-E35) (HR, 7.27; 95% CI, 6.19-8.54) followed by overweight, obesity, and other hyperalimentation (E65-E68) (HR, 2.33; 95% CI, 2.02-2.68). The remaining disease subgroups within the chapter of endocrine diseases also showed higher risks (HR, 1.03-1.54), though not all reached statistical significance. These results appeared even more elevated for overall gynecomastia.

Table 2.

Hazard ratios of incident disease (by ICD chapters) during follow-up after a diagnosis of gynecomastia (idiopathic^a, overall) compared with reference males

Disease chapter	ICD-10 code	ICD-8 code	Disease events (n)b		Disease nonevents (n)c		Hazard ratio (95% CI)
			Idiopathica	Overall	Idiopathica	Overall	Idiopathica	Overall
Infectious	A00-A99	000-136	3111	5081	47 877	66 817	1.34 (1.23-1.47)	1.55 (1.45-1.66)
Cancer	C00-D48	140-239d	3312	5050	74 484	94 652	1.26 (1.15-1.37)	1.37 (1.27-1.47)
Blood	D50-D64	280-289	1778	3251	83 626	115 795	1.29 (1.15-1.46)	1.72 (1.58-1.88)
Endocrine	E00-E90	240-279	4113	5772	54 369	65 496	1.89 (1.76-2.03)	2.00 (1.88-2.13)
Psychiatric	F00-F99	290-315	4703	6365	56 557	74 107	1.33 (1.24-1.43)	1.45 (1.36-1.54)
Neurological	G00-H59	320-379	3696	5061	25 614	31 101	1.30 (1.20-1.41)	1.41 (1.31-1.51)
Circulatory	I00-I99	390-458	3553	4489	51 677	61 037	1.27 (1.17-1.39)	1.41 (1.31-1.52)
Pulmonary	J00-J99	460-519	2450	3679	15 196	20 915	1.19 (1.07-1.32)	1.34 (1.23-1.45)
Gastrointestinal	K00-K93	520-577	2573	3145	15 949	18 641	1.35 (1.23-1.49)	1.41 (1.30-1.55)
Skin	L00-L99	680-709	3543	4951	42 927	58 877	1.42 (1.31-1.54)	1.61 (1.50-1.72)
Musculoskeletal and connective tissue	M00-M99	710-738	3656	4366	14 080	16 304	1.55 (1.43-1.68)	1.57 (1.45-1.69)
Congenital malformations	Q00-Q99	740-759	832	1235	58 880	88 279	1.35 (1.14-1.60)	1.62 (1.41-1.85)

Abbreviation: ICD, International Classification of Diseases.

This group of males did not have a known risk factor before the gynecomastia diagnosis.

Total frequency of incident events within the disease group among males (with gynecomastia and their age- and calendar-matched references) within each stratum.

Total frequency of nonevents within the disease group for males (with gynecomastia and their age- and calendar-matched references) within each stratum.

For the Danish Cancer Registry, ICD-7 codes: 140-239 were used instead of ICD-8 codes.

Figure 1.

Hazard ratios of incident disease (by ICD chapters) during follow-up of a diagnosis of gynecomastia (idiopathic, overall) compared with their reference males.

Disease risks as a function of age at the time of gynecomastia are presented in Supplementary Fig. S1 (21). For infectious and psychiatric disease, age at the time of gynecomastia diagnosis was central in determining HRs and a peak in HRs was observed in age ranges 30 to 50 years for both disease chapters. For endocrine diseases, the highest HRs were observed in the age range 20 to 40 years, after which HRs declined with age at time of diagnosis. These observations were evident for both idiopathic and overall gynecomastia, but noteworthy is the higher HRs across all age categories for overall gynecomastia.

Disease risks following gynecomastia with a known risk factor were statistically significant elevated across all included disease chapters (HR, 1.64-2.42), with the highest incident disease risks for diseases of the blood (HR, 2.42; 95% CI, 2.14-2.74) and endocrine disorders (HR, 2.34; 95% CI, 2.08-2.64) (Supplementary Table S5) (21).

Morbidities Before Gynecomastia Diagnosis

The associations between morbidities diagnosed before/on the date of gynecomastia diagnosis are presented in Table 3 and visualized in Fig. 2. Idiopathic gynecomastia was mostly associated with statistically significant higher odds of previous disease across all included ICD disease chapters (OR, 1.05-1.51), except for psychiatric diseases (OR, 0.72; 95% CI, 0.68-0.78), compared with males without gynecomastia. The most prevalent prior incident diseases among males with idiopathic gynecomastia were musculoskeletal and connective tissue (OR, 1.51; 95% CI, 1.46-1.57) and circulatory (OR, 1.36; 95% CI, 1.29-1.43) diseases (Table 3).

Table 3.

Odds ratios of comorbidities (by ICD chapters) diagnosed before/on the date of gynecomastia diagnosis (idiopathic^a, overall) compared with reference males

Disease chapter	ICD-10 code	ICD-8 code	Disease events before diagnosis (n)b		Disease nonevents before diagnosis (n)c		Odds ratio (95% CI)
			Idiopathica	Overall	Idiopathica	Overall	Idiopathica	Overall
Infectious	A00-A99	000-136	10 214	15 121	87 304	125 453	1.20 (1.14-1.27)	1.52 (1.46-1.59)
Cancer	C00-D48	140-239d	5004	11 135	92 514	129 439	1.19 (1.10-1.29)	1.88 (1.79-1.98)
Blood	D50-D64	280-289	2177	3997	95 341	136 577	1.20 (1.08-1.34)	1.73 (1.61-1.87)
Endocrine	E00-E90	240-279	9225	17 886	88 293	122 688	1.05 (0.99-1.11)	1.98 (1.90-2.06)
Psychiatric	F00-F99	290-315	7441	12 704	90 077	127 870	0.72 (0.68-0.78)	1.66 (1.59-1.74)
Neurological	G00-H59	320-379	20 162	33 702	77 356	106 872	1.20 (1.15-1.25)	1.48 (1.43-1.53)
Circulatory	I00-I99	390-458	14 714	29 461	82 804	111 113	1.36 (1.29-1.43)	2.01 (1.94-2.09)
Pulmonary	J00-J99	460-519	25 316	36 791	72 202	103 783	1.26 (1.21-1.31)	1.52 (1.48-1.57)
Gastrointestinal	K00-K93	520-577	26 778	44 027	70 740	96 547	1.26 (1.21-1.31)	1.61 (1.56-1.66)
Skin	L00-L99	680-709	11 658	17 883	85 860	122 691	1.30 (1.24-1.37)	1.52 (1.46-1.58)
Musculoskeletal and connective tissue	M00-M99	710-738	30 018	48 552	67 500	92 022	1.51 (1.46-1.57)	1.77 (1.72-1.83)
Congenital malformations	Q00-Q99	740-759	8027	10 765	89 491	129 809	1.10 (1.03-1.16)	1.33 (1.27-1.40)

Abbreviation: ICD, International Classification of Diseases.

This group of males did not have a known risk factor before the gynecomastia diagnosis.

Total frequency of incident events within the disease group amongst males (with gynecomastia and their age- and calendar-matched references) within each stratum.

Total frequency of nonevents within the disease group for males (with gynecomastia and their age- and calendar-matched references) within each stratum.

For the Danish Cancer Registry, ICD-7 codes: 140-239 were used instead of ICD-8 codes.

Figure 2.

Odds ratios of comorbidities (by ICD chapters) diagnosed before or on the date of gynecomastia diagnosis (idiopathic, overall) compared with their reference males.

Previous diseases according to gynecomastia with a known risk factor were highly statistically significant across all included ICD-disease chapters (OR, 2.00-4.18), especially endocrine (OR, 4.04; 95% CI, 3.81-4.29), circulatory (OR, 3.48; 95% CI, 3.27-3.71), and psychiatric (OR, 4.18; 95% CI, 3.92-4.45) diseases (Supplementary Table S6) (21).

Discussion

In this unique study with access to nationwide register-based information on known underlying causes of gynecomastia, we found that idiopathic gynecomastia was associated with a statistically significant higher risk of future disease across all included ICD-10 disease chapters, especially for endocrine diseases. A male diagnosed with idiopathic gynecomastia also presented with more morbidities before diagnosis for most disease chapters, expect for psychiatric disorders, compared with males with no gynecomastia. The associations were even more elevated for males with a known risk factor to the gynecomastia diagnosis.

To our knowledge, this is the first study comprehensively investigating past and future health risks associated with gynecomastia across ICD disease chapters. Although gynecomastia is not considered a severe disease, we clearly demonstrate a marked increased morbidity among males diagnosed with gynecomastia, both related to previous and future health. This may be expected for gynecomastia cases identified with a known risk factor linked to underlying pathologic conditions, but the observed health effects associated with idiopathic gynecomastia are largely surprising. Although the health risks were more elevated for gynecomastia with a known risk factor, we provide significant evidence into a higher disease risks for both males with and without a known underlying pathology. These chapter-specific associations were statistically significant across all disease chapter, providing HRs of up to 2.00. The most significant identified future health risk was endocrine diseases, which is in accordance with the known risk of gynecomastia following an unfavorable circulating E2/T ratio, as seen in hypogonadism and differences in sex development (DSD) including sex chromosome disorders, increased thyroid hormone levels/(thyrotoxicosis), or growth hormone therapy in elderly growth hormone-deficient males. Intriguingly, few subgroups of diseases such as disorders of endocrine glands resulted in very high and statistically significant risk estimates, indicating strong associations. Altogether, this emphasizes the interaction between gynecomastia and other conditions with endocrine origins. The noteworthy prior morbidities of musculoskeletal and connective tissue and circulatory diseases further recognize that gynecomastia may not merely be perceived as an isolated condition but is closely linked to more severe diseases. In addition to previously identified mental and psychosocial consequences of gynecomastia (12, 23–25), our study findings further stress the significance of other associated morbidities, regardless of underlying pathology. The negative association with psychiatric disorders before idiopathic gynecomastia, but not for overall gynecomastia, is difficult to explain but may be due to a relatively large group of psychiatric patients who do not have the capacity to seek further medical assessment for other mild diseases such as gynecomastia, whereas on the other hand, males diagnosed with another condition that causes gynecomastia would have more thorough clinical investigation, including a psychiatric/gynecomastia diagnoses. Consequently, this may lead to potential underreporting of gynecomastia in the idiopathic group of patients.

Despite our study not establishing any causality of an association, our study suggests that gynecomastia is strongly interlinked with other endocrine diseases and may be a first clinical symptom of future disease. However, we acknowledge that gynecomastia does not represent an isolated cause to future morbidities and that males can be predisposed to morbidities before the manifestation of gynecomastia, whereby other associated diseases may be the true predictor of a future disease risk.

Gynecomastia is a complex phenomenon, and the true prevalence remains unclear. Existing studies have each reported varying prevalence of gynecomastia, which is most likely from differences in the definition of gynecomastia. Interestingly, Klang and coworkers examined quantitative computed tomography scans performed on males with gynecomastia from the general population (26) and they found a large variation in male breast tissue diameters (22-36 mm corresponding to 90th-97.5th cumulative percentiles of diameters). They also reported highest prevalence of gynecomastia (>22 mm) in 13- to 14-year-old boys (37%) and males in the ninth decade (21%) (26). In the only existing report on incidence of gynecomastia, our research group showed a marked increasing incidence of gynecomastia in Denmark from 1998 through 2017 (10), and the associated adverse health effects of gynecomastia are therefore of significant concern. Our results clearly illustrate the importance of early diagnosis, full clinical examination, and follow-up of these males to identify underlying pathologies for effective management and prevention of future diseases. More specifically, the need for aiding treatment options and possible interventions with focus on diet and lifestyle may be relevant to elucidate. We identified use of certain medications as the most frequent pathologic source to gynecomastia, thus targeting interventions to certain subpopulations of males with gynecomastia could be considered. One example may be the bodybuilding environment, or rather young individuals who simply use anabolic androgen steroids to look leaner and muscular (27). In addition, alcohol addiction and obesity with no specified endocrine origin were identified as prevalent known pathologies, which could also be targeted in interventions with focus on lifestyle, diet and education, albeit any intervention is difficult after diagnosis. Importantly, addressing this issue is complex given the different characteristics in underlying pathology by age, and any relevant interventions need to account for identified age disparities.

Strengths and Limitations of Our Study

Our study has several strengths. We used data from nationwide registries, known for quality and completeness, enabling a large study population with extensive and valid information on diseases collected for several decades. The large study also provided sufficient statistical power and enabled an isolated focus on idiopathic cases without influence of underlying pathologies. Because health care is free in Denmark, the gynecomastia ascertainment is expected to be reliable with limited bias because of socioeconomic factors. The population-based study design provided clinically important associations between gynecomastia and related comorbidities and diseases risks; however, these remain indications and do not estimate any causal effects.

A potential source of bias is that patients with gynecomastia are already in contact with the health care system and may be more likely than males without gynecomastia to be diagnosed with other diseases in the process. Because gynecomastia in most cases is a relatively mild condition, we cannot exclude some misclassification related to males who were not seeking contact with the hospital, or males who went to private facilities for cosmetic treatment without a diagnosis in the registries. As with registry-based studies, we had no control of the quality in making a gynecomastia diagnosis and use of correct guidelines, which potentially risk misclassification of pseudo-gynecomastia (1). The identification of gynecomastia with a known risk factor was based solely on existing knowledge and anecdotal reports and we acknowledge that some unidentified or not yet recognized risk factors may also play a role in the development in gynecomastia. In addition, the inclusion of cannabis as a risk factor remains controversial; however, this only involved 6 cases and did not affect the given results.

The increasing incidence of gynecomastia in Denmark during the past 20 years imply a changed endogenous or exogenous sex-steroid environment, perhaps in addition to increasing obesity over time (10). However, we were not able to adjust for potential confounders (apart from age and calendar time by design) such as obesity, exposure to endocrine disrupting chemicals, or steroid use that may be important in the relationship between gynecomastia and disease risk. Because obesity is known to be a risk factor of several groups of diseases and further associated with gynecomastia, we recognize that obesity may explain some of our findings, representing a causal cause.

Conclusions

In our nationwide register-based study, idiopathic gynecomastia was associated with a marked increased risk of future diseases. The observed associations were even more significant for males identified with a known preexisting risk factor before gynecomastia. This demonstrates the potential importance of gynecomastia as an important first clinical symptom of an underlying disease and a significant predictor of future disease risk. We hope our present findings will stimulate more awareness among health care providers to increase identification of gynecomastia and its causes in males. Further studies are needed to establish the specific long-term health consequences of gynecomastia.

Abbreviations

E2

estradiol

HR

hazard ratio

ICD-10

International Classification of Diseases, 10th revision

OR

odds ratio

Funding

The salaries of A.J., E.V.B., and C.S.U. were partially supported by a grant from the National Institutes of Health (Grant no. 1R01CA236816-01A1). E.V.B. and Y.H.L. were also partially supported by grants from the Danish Health Foundation, Helsefonden (Grant no. 21-B-0063). No other disclosures are declared.

Author Contributions

A.J. and E.V.B. contributed to the concept and design of the study. E.V.B. performed data cleanup and prepared data. Y.H.L. performed statistical analyses and created figures. All authors contributed to critical interpretation of data and results. C.S.U. drafted the original manuscript with substantial contributions from E.V.B. All authors reviewed and finally approved the manuscript.

Disclosures

The authors have nothing to disclose.

Data Availability

Data used in this study are governed and maintained centrally by the Danish Data Health Authority and data access is regulated by EU General Data Protection Regulations (GPDR). Data are not publicly available and pseudo-anonymized data can only be accessed after approval by the Danish Data Health Authority and the Danish Data Protection Agency.