Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Association of Androgenetic Alopecia With Smoking and Its Prevalence Among Asian Men: A Community-Based Survey

Association of Androgenetic Alopecia With Smoking and Its Prevalence Among Asian Men: A... Abstract Objectives To evaluate the association of androgenetic alopecia (AGA) with smoking and to estimate its prevalence among Asian men. Design Population-based cross-sectional survey. Setting Tainan County, Taiwan. Participants The eligible population consisted of all male residents 40 years or older in Tainan County. A total of 740 subjects aged 40 to 91 years participated in the survey between April 10, 2005, and June 12, 2005. Main Outcome Measures Norwood and Ludwig classifications were used to assess the degree of hair loss. Information on smoking, together with other possible risk factors and age at onset of AGA, was collected using a questionnaire interview. Results After controlling for age and family history, statistically significant positive associations were noted between moderate or severe AGA (Norwood types ≥IV) and smoking status (odds ratio [OR], 1.77; 95% confidence interval [CI], 1.14-2.76), current cigarette smoking of 20 cigarettes or more per day (OR, 2.34; 95% CI, 1.19-4.59), and smoking intensity (OR, 1.78; 95% CI, 1.03-3.07). The OR of early-onset history for AGA grades increased in a dose-response pattern. Risk for moderate or severe AGA increased for family history of first-degree and second-degree relatives, as well as for paternal relatives. Conclusions The age-specific prevalence of AGA in Taiwan was compatible to that among Korean men but was lower than that among persons of white race/ethnicity. Smoking status, current amount of cigarette smoking, and smoking intensity were statistically significant factors responsible for AGA after controlling for age and family history. Patients with early-onset AGA should receive advice early to prevent more advanced progression. Androgenetic alopecia (AGA), a hereditary androgen-dependent disorder, is characterized by progressive thinning of the scalp hair defined by various patterns.1 It is the most common type of hair loss in men.2 Lower prevalence has been seen among Asian, Native American, and African American men, whereas high prevalence has been found among men of white race/ethnicity.3 The changes of AGA are androgen dependent and follow an inheritance mode with gene polymorphisms.3 To our knowledge, only 1 gene that encodes the androgen receptor has been identified.4 While prerequisites are androgens and a genetic predisposition, clinical practice has shown that simply blocking androgens does not result in the conversion of miniaturized follicles to terminal ones in advanced alopecia.1 Some environmental factors such as smoking may play a role in the pathogenesis of AGA. The association between smoking and AGA has been addressed in 3 studies5-7 with inconsistent results. One study5 showed a positive association, one study6 failed to demonstrate a statistically significant positive association, and another study7 showed the opposite findings, albeit statistically nonsignificant. The first objective of this study was to estimate the prevalence and types of AGA among Taiwanese men and to compare our findings with those in other countries. The second objective was to investigate the association between smoking, family history, and other potential risk factors and AGA. Methods Study subjects and design Eligible subjects in this study consisted of male residents 40 years or older in Tainan County, Taiwan. Our study was part of a community-based integrated screening program that invited residents 30 years or older. The details of a screening program similar to ours in Tainan County have been published elsewhere.8 A community-based survey on prevalence of AGA was added to this integrated community-based screening program between April 10, 2005, and June 12, 2005, by inviting 929 residents who were selected from a population household registry of Tainan County in light of different screening criteria based on evidence-based medicine. Most participants were 40 years or older; this age group is subsidized by the local government. Only 5 subjects were aged 30 to 39 years, and they were excluded from the study. Of the remaining 924 subjects invited, the overall response rate to participate in our survey on AGA was 80.1% (740 of 924). The age-specific response rates were 83.9% for those aged 40 to 49 years (99 of 118), 81.9% for those aged 50 to 59 years (95 of 116), 84.9% for those aged 60 to 69 years (248 of 292), and 74.9% for those aged 70 years and older (298 of 398). There was a similar distribution with respect to age between respondents and nonrespondents. Therefore, the study population for age-specific prevalence in the AGA survey is representative of the underlying population invited to attend the screening. The study protocol was approved by the ethics committee of the Bureau of Tainan County, and informed consent was obtained from all participants. Classification of aga The diagnosis of AGA was based on the pattern of hair loss of the participants. Assessment of the degree of hair loss was obtained by independent public health nurses trained by a dermatologist. To classify the degree of AGA for each subject, the classification by Norwood,2 a standard classification scheme with good test-retest reliability, was used. Some men with “female pattern hair loss” (noninvolvement of the frontal hairline) were assessed separately using the classification by Ludwig.9 Androgenetic alopecia among men with cosmetically significant male pattern baldness was defined as type III or greater (types III [including IIIv and IIIa], IV [including IVa], V [including Va], VI, and VII) according to the Norwood classification. The prevalence of more advanced degrees of alopecia characterized by only a remaining horseshoe fringe of hair (Norwood types V, Va, VI, and VII) was also estimated. The prevalence of AGA with female pattern hair loss was reported separately. Norwood type IV of AGA represents the starting grade of severe frontal AGA concurrent with vertex AGA. Therefore, we divided AGA into 2 categories (mild AGA [Norwood types I-III] and moderate or severe AGA [Norwood types IV-VII]) to assess its association with smoking and other potential risk factors. Smoking and other possible risk factors In addition to classification of AGA, we collected information on age at onset of AGA together with smoking and other possible risk factors using a face-to-face questionnaire interview. Smoking status (never, quit, or current); age at start of smoking; and quantity, duration, and frequency of smoking were collected using a structured questionnaire administered by trained staff members. In addition to smoking status, we categorized cigarette smoking into 4 groups (never, quit, current smoker of <20 cigarettes per day, or current smoker of ≥20 cigarettes per day) for comparison. Regarding anthropometric measures, body weight, body height, waist circumference, and hip circumference were measured. Body mass index was calculated as weight in kilograms divided by height in meters squared. Blood pressure was measured twice with at least a 5-minute interval between measurements. Hypertension was defined as a systolic blood pressure of 140 mm Hg or higher and a diastolic blood pressure of 90 mm Hg or higher or as taking antihypertensive medication. To collect biochemical markers, a venous blood sample was obtained after 12-hour fasting to check blood glucose, triglycerides, total cholesterol, and high-density lipoprotein cholesterol levels. Low-density lipoprotein cholesterol level was calculated in accord with the following formula: low-density lipoprotein cholesterol level = [(total cholesterol level − high-density lipoprotein cholesterol level − (triglyceride level/5)]. If the triglyceride level was greater than 400 mg/dL (to convert to millimoles per liter, multiply by 0.0113), the low-density lipoprotein cholesterol level was set as a missing value. Abnormal status for these biochemical markers included hypertriglyceridemia (triglyceride level, >160 mg/dL), hypercholesterolemia (total cholesterol level, >220 mg/dL [to convert to millimoles per liter, multiply by 0.0259]), low high-density lipoprotein cholesterol level (<42 mg/dL), high low-density lipoprotein cholesterol level (≥140 mg/dL), and high fasting glucose level (>110 mg/dL [to convert to millimoles per liter, multiply by 0.0555]). Dyslipidemia was defined as abnormal serum triglyceride, total cholesterol, high-density lipoprotein cholesterol, or low-density lipoprotein cholesterol level or as taking lipid-lowering drugs. In each subject, a questionnaire regarding diagnoses of chronic diseases such as hypertension, diabetes mellitus, cardiovascular disease, antihypertensive and lipid-lowering drug use, timing of growth spurt in puberty, socioeconomic factors, alcoholic beverage consumption, and betel nut chewing history was completed. The socioeconomic factors included age, occupation, and level of education. Age was defined as the age at study enrollment. Information about baldness among the first-degree, second-degree, and third-degree relatives was also collected using a structured questionnaire administered by trained staff members. The degree of hair loss was recalled by the study subjects. Statistical analysis The age-specific prevalence of AGA (Norwood types ≥III) was expressed as a percentage. A multivariate logistic regression model was used to assess the relationships of each possible risk factor such as smoking and degrees of relative relationships for family history of AGA (first-degree, second-degree, and third-degree relatives) to moderate or severe AGA (Norwood types ≥IV). A cumulative logits model was used to assess the incremental effect of age at onset of AGA (≤40 vs >40 years) on AGA severity (Norwood types I-VII) with ordinal properties. The proportional odds assumption was checked for the 7 AGA grades. In addition, the association between family history of AGA and age at onset of AGA was assessed. Because age and family history of AGA were 2 established factors for AGA, they were retained in the model in the multivariate analysis. All odds ratios (ORs) and their 95% confidence intervals (CIs) were computed using statistical software (SAS version 8; SAS Institute, Inc, Cary, North Carolina). Results Of 924 invited men, 740 participated in this survey. The mean ± SD age of participants was 65.2 ± 11.2 (age range, 40-91 years). Prevalence The age-specific number and percentage of different AGA types are given in Table 1Figure 1 shows that the age-specific prevalence of AGA (Norwood types III [including IIIv and IIIa], IV [including IVa], V [including Va], VI, and VII) increased with advancing age. The age-specific prevalence of severe hair loss (Norwood types V, Va, VI, and VII) was low. In Figure 2, the age-specific prevalence of AGA (Norwood types ≥III) is compared with the results of previous studies.2,7,10-15 Of 740 men examined, 48 (6.5%) were Norwood type A variants, higher than that reported in the study by Norwood (3%), and 13 (1.8%) manifested female pattern AGA. The prevalence did not increase with age for Norwood type A variants or for Ludwig grades L-I to L-III. For the groups aged 40 to 49, 50 to 59, 60 to 69, and 70 years or older, the age-specific prevalences of Norwood type A variants were 5.1%, 1.1%, 8.1%, and 7.4%, respectively, and those of female pattern AGA were 0.0%, 2.1%, 1.6%, and 2.4%, respectively (Table 1). Association In the univariate analysis, smokers were at increased risk of having moderate or severe AGA (Norwood types ≥IV) (OR, 1.61; 95% CI, 1.05-2.46). However, no statistically significant associations were found for intensity or duration of smoking (Table 2). There were statistically significant positive associations between moderate or severe AGA and smoking status (OR, 1.77; 95% CI, 1.14-2.76), current cigarette smoking of 20 cigarettes or more per day (OR, 2.34; 95% CI, 1.19-4.59), smoking intensity (OR, 1.78; 95% CI, 1.03-3.07), and dyslipidemia (OR, 1.47; 95% CI, 1.01-2.14) after adjusting for age and family history of AGA (Table 2). Smoking status and other statistically significant variables (age, family history of AGA, dyslipidemia, and betel nut chewing) were retained in the model and were adjusted for each other. A statistically significant association between smoking status and moderate or severe AGA (OR, 1.63; 95% CI, 1.00-2.65) still remained. Early-onset aga and aga grades The incremental effect of early-onset on severity of AGA was demonstrated by using the cumulative logits model. Findings suggest that earlier onset of AGA was associated with severe AGA. In those with early-onset AGA, AGA grades were statistically significantly worse (OR, 2.09; 95% CI, 1.14-3.85). Early-onset AGA was also related to family history of AGA (OR, 2.89; 95% CI, 1.42-5.88) using the logistic regression model. Risk for moderate or severe AGA increased with degree of relationship for family history (first degree [OR, 13.38; 95% CI, 4.80-37.27], second degree [OR, 6.33; 95% CI, 2.37-16.91], and third degree [OR, 5.32; 95% CI, 0.80-35.48]) (Table 3). The association between moderate or severe AGA and family history of AGA from paternal relatives was statistically significant (OR, 12.69; 95% CI, 4.65-34.60), whereas the corresponding association from maternal relatives was not (OR, 3.07; 95% CI, 0.30-31.32). The OR for the association between family history in both paternal and maternal relatives could not be estimated because there was no case with a positive family history from both paternal and maternal relatives among the subjects without AGA. Comment In our study, the age-specific prevalence of male pattern AGA was comparable with that among Korean men.12 Compared with the results of another Asian study15 conducted in Singapore, our prevalence rate of AGA was lower. However, the comparison should be interpreted with caution because the definition of cosmetically significant AGA by Tang et al15 was Norwood type I or greater, which was different from the definition (Norwood types ≥III) used in our study. This may also explain why the age-specific prevalence of AGA in their study was much higher than that in our study. The age-specific prevalence of AGA in another Asian study13 conducted in Thailand was also higher than that in our study. However, race/ethnicity in Taiwan is different from that in Thailand. The prevalence of AGA increased steadily with advancing age but was lower than that among persons of white race/ethnicity. An increase in risk of AGA with age reflects the natural progression of this condition. Advancing age is a risk factor for AGA in all men, irrespective of their family history. In addition, female pattern hair loss was observed in 13 of 740 men (1.8%) examined. Because AGA in some men may follow the female (Ludwig9) pattern, we suggest that the female pattern should be included in the classification of AGA in men. A positive association between smoking and AGA was demonstrated in our study. Our results were consistent with findings by Mosley and Gibbs5 using a cross-sectional survey in a general surgical outpatient clinic in the United Kingdom. Unlike our study that was based on the general population, the generalizability of their study may be limited because respondents came from a general surgical outpatient clinic. In addition, no confounders were considered, and the dose-response relationship to AGA with respect to amount and duration of smoking was not elucidated. Our positive results were not comparable with the findings from 2 studies.6,7 Severi et al7 reported a lower risk of AGA among current smokers (OR, 0.86; 95% CI, 0.54-1.38) and exsmokers (OR, 0.91; 95% CI, 0.65-1.29), although the results were not statistically significant. Nevertheless, details of smoking history were lacking, and the age at onset of AGA was unknown. In addition, different AGA classification categories instead of the Norwood classification were used. Therefore, it was difficult to compare their results with those of other studies. Matilainen et al6 found that the prevalence of current smoking among women with extensive hair loss (Ludwig grades ≥L-II to L-III) did not differ from those with normal or minimal hair loss (P =.92). However, the main objective of the study was not tailored to smoking history, and no definition of smoking was addressed. The mechanisms by which smoking causes hair loss may be multifactorial.1 First, cigarette smoking may be deleterious to the microvasculature of the dermal hair papilla. Second, smoke genotoxicants may do damage to DNA of the hair follicle. Third, smoking may lead to an imbalance in the follicular protease or antiprotease system. Smoking-induced oxidative stress may lead to the release of proinflammatory cytokines that, in turn, results in follicular microinflammation and fibrosis. Fourth, cigarette smoking may yield a relative hypoestrogenic state by inducing increased hydroxylation of estradiol and inhibition of aromatase.1 Although the genetic basis of AGA is well documented in the medical community and among the general population, there are few studies investigating the familial aggregation of AGA. Chumlea et al16 reported that the risk of AGA increased among men with a positive maternal grandfather history and even more with a history of hair loss in their father. In our study, moderate or severe AGA was statistically significantly associated with family history of AGA among the first-degree and second-degree relatives but not among the third-degree relatives after adjusting for age. The highest OR was associated with a family history among first-degree relatives. In addition, family history of AGA among paternal relatives was statistically significantly predictive of moderate or severe AGA after adjusting for age. Our findings do not support an association between moderate or severe AGA and family history of AGA among maternal relatives. The effect of family history among both paternal and maternal older relatives could not be assessed because no case was found among subjects without AGA. Moreover, we found that family history of AGA is statistically significantly associated with the risk of early-onset AGA. This implies that those with a family history of AGA may have a higher risk of early-onset AGA and a higher risk of developing severe AGA. Most important, early-onset AGA (age at onset ≤40 years) showed a statistically significant dose-dependent association with AGA grade after adjusting for age and family history. From the clinical point of view, this suggests that patients with early-onset AGA should receive early advice to prevent the deterioration of AGA. In conclusion, our study showed that the prevalence of AGA among men in Taiwan was lower than that among persons of white race/ethnicity. The severity of AGA was also milder in Taiwan. We suggest that a female pattern should be included in the classification because AGA in some men may follow this Ludwig pattern. We also confirmed that smoking status, current amount of cigarette smoking, and smoking intensity play important roles in the development of moderate or severe AGA. Correspondence: Tony Hsiu-Hsi Chen, DDS, PhD, Division of Biostatistics, Institute of Preventive Medicine, College of Public Health, National Taiwan University, Room 521, Fifth Floor, 17 Suchow Rd, Taipei 100, Taiwan (chenlin@ntu.edu.tw). Financial Disclosure: None reported. Back to top Article Information Accepted for Publication: May 3, 2007. Author Contributions: Drs Su and Chen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Chen. Acquisition of data: Su. Analysis and interpretation of data: Chen and Su. Drafting of the manuscript: Su. Critical revision of the manuscript for important intellectual content: Chen. Statistical analysis: Chen. Administrative, technical, and material support: Chen. Study supervision: Chen. Additional Contributions: Yao-Der Chen, MS, and other colleagues helped with data collection, and Min-Shu Hsu, BSc, performed data management. References 1. Trüeb RM Molecular mechanisms of androgenetic alopecia. Exp Gerontol 2002;37 (8-9) 981- 990PubMedGoogle ScholarCrossref 2. Norwood OT Male pattern baldness: classification and incidence. South Med J 1975;68 (11) 1359- 1365PubMedGoogle ScholarCrossref 3. Hoffmann R Male androgenetic alopecia. Clin Exp Dermatol 2002;27 (5) 373- 382PubMedGoogle ScholarCrossref 4. Ellis JASinclair RHarrap SB Androgenetic alopecia: pathogenesis and potential for therapy. Expert Rev Mol Med 2002;20021- 11PubMedGoogle Scholar 5. Mosley JGGibbs AC Premature grey hair and hair loss among smokers: a new opportunity for health education? BMJ 1996;313 (7072) 1616PubMedGoogle ScholarCrossref 6. Matilainen VLaakso MHirsso PKoskela PRajala UKeinanen-Kiukaanniemi S Hair loss, insulin resistance, and heredity in middle-aged women: a population-based study. J Cardiovasc Risk 2003;10 (3) 227- 231PubMedGoogle ScholarCrossref 7. Severi GSinclair RHopper JL et al. Androgenetic alopecia in men aged 40-69 years: prevalence and risk factors. Br J Dermatol 2003;149 (6) 1207- 1213PubMedGoogle ScholarCrossref 8. Chen THChiu YHLuh DL et al. Taiwan Community-Based Integrated Screening Group, Community-based multiple screening model: design, implementation, and analysis of 42,387 participants. Cancer 2004;100 (8) 1734- 1743PubMedGoogle ScholarCrossref 9. Ludwig E Classification of the types of androgenetic alopecia (common baldness) occurring in the female sex. Br J Dermatol 1977;97 (3) 247- 254PubMedGoogle ScholarCrossref 10. DeMuro-Mercon CRhodes TGirman CJVatten L Male-pattern hair loss in Norwegian men: a community-based study. Dermatology 2000;200 (3) 219- 222PubMedGoogle ScholarCrossref 11. Hamilton JB Patterned loss of hair in man: types and incidence. Ann N Y Acad Sci 1951;53 (3) 708- 728PubMedGoogle ScholarCrossref 12. Paik JHYoon JBSim WYKim BSKim NI The prevalence and types of androgenetic alopecia in Korean men and women. Br J Dermatol 2001;145 (1) 95- 99PubMedGoogle ScholarCrossref 13. Pathomvanich DPongratananukul SThienthaworn PManoshai S A random study of Asian male androgenetic alopecia in Bangkok, Thailand. Dermatol Surg 2002;28 (9) 804- 807PubMedGoogle Scholar 14. Rhodes TGirman CJSavin RC et al. Prevalence of male pattern hair loss in 18-49 year old men. Dermatol Surg 1998;24 (12) 1330- 1332PubMedGoogle ScholarCrossref 15. Tang PHChia HPCheong LLKoh D A community study of male androgenetic alopecia in Bishan, Singapore. Singapore Med J 2000;41 (5) 202- 205PubMedGoogle Scholar 16. Chumlea WCRhodes TGirman CJ et al. Family history and risk of hair loss. Dermatology 2004;209 (1) 33- 39PubMedGoogle ScholarCrossref http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Archives of Dermatology American Medical Association

Association of Androgenetic Alopecia With Smoking and Its Prevalence Among Asian Men: A Community-Based Survey

Archives of Dermatology , Volume 143 (11) – Nov 1, 2007

Loading next page...
 
/lp/american-medical-association/association-of-androgenetic-alopecia-with-smoking-and-its-prevalence-NxVXuyfpyM

References (18)

Publisher
American Medical Association
Copyright
Copyright © 2007 American Medical Association. All Rights Reserved.
ISSN
0003-987X
eISSN
1538-3652
DOI
10.1001/archderm.143.11.1401
pmid
18025364
Publisher site
See Article on Publisher Site

Abstract

Abstract Objectives To evaluate the association of androgenetic alopecia (AGA) with smoking and to estimate its prevalence among Asian men. Design Population-based cross-sectional survey. Setting Tainan County, Taiwan. Participants The eligible population consisted of all male residents 40 years or older in Tainan County. A total of 740 subjects aged 40 to 91 years participated in the survey between April 10, 2005, and June 12, 2005. Main Outcome Measures Norwood and Ludwig classifications were used to assess the degree of hair loss. Information on smoking, together with other possible risk factors and age at onset of AGA, was collected using a questionnaire interview. Results After controlling for age and family history, statistically significant positive associations were noted between moderate or severe AGA (Norwood types ≥IV) and smoking status (odds ratio [OR], 1.77; 95% confidence interval [CI], 1.14-2.76), current cigarette smoking of 20 cigarettes or more per day (OR, 2.34; 95% CI, 1.19-4.59), and smoking intensity (OR, 1.78; 95% CI, 1.03-3.07). The OR of early-onset history for AGA grades increased in a dose-response pattern. Risk for moderate or severe AGA increased for family history of first-degree and second-degree relatives, as well as for paternal relatives. Conclusions The age-specific prevalence of AGA in Taiwan was compatible to that among Korean men but was lower than that among persons of white race/ethnicity. Smoking status, current amount of cigarette smoking, and smoking intensity were statistically significant factors responsible for AGA after controlling for age and family history. Patients with early-onset AGA should receive advice early to prevent more advanced progression. Androgenetic alopecia (AGA), a hereditary androgen-dependent disorder, is characterized by progressive thinning of the scalp hair defined by various patterns.1 It is the most common type of hair loss in men.2 Lower prevalence has been seen among Asian, Native American, and African American men, whereas high prevalence has been found among men of white race/ethnicity.3 The changes of AGA are androgen dependent and follow an inheritance mode with gene polymorphisms.3 To our knowledge, only 1 gene that encodes the androgen receptor has been identified.4 While prerequisites are androgens and a genetic predisposition, clinical practice has shown that simply blocking androgens does not result in the conversion of miniaturized follicles to terminal ones in advanced alopecia.1 Some environmental factors such as smoking may play a role in the pathogenesis of AGA. The association between smoking and AGA has been addressed in 3 studies5-7 with inconsistent results. One study5 showed a positive association, one study6 failed to demonstrate a statistically significant positive association, and another study7 showed the opposite findings, albeit statistically nonsignificant. The first objective of this study was to estimate the prevalence and types of AGA among Taiwanese men and to compare our findings with those in other countries. The second objective was to investigate the association between smoking, family history, and other potential risk factors and AGA. Methods Study subjects and design Eligible subjects in this study consisted of male residents 40 years or older in Tainan County, Taiwan. Our study was part of a community-based integrated screening program that invited residents 30 years or older. The details of a screening program similar to ours in Tainan County have been published elsewhere.8 A community-based survey on prevalence of AGA was added to this integrated community-based screening program between April 10, 2005, and June 12, 2005, by inviting 929 residents who were selected from a population household registry of Tainan County in light of different screening criteria based on evidence-based medicine. Most participants were 40 years or older; this age group is subsidized by the local government. Only 5 subjects were aged 30 to 39 years, and they were excluded from the study. Of the remaining 924 subjects invited, the overall response rate to participate in our survey on AGA was 80.1% (740 of 924). The age-specific response rates were 83.9% for those aged 40 to 49 years (99 of 118), 81.9% for those aged 50 to 59 years (95 of 116), 84.9% for those aged 60 to 69 years (248 of 292), and 74.9% for those aged 70 years and older (298 of 398). There was a similar distribution with respect to age between respondents and nonrespondents. Therefore, the study population for age-specific prevalence in the AGA survey is representative of the underlying population invited to attend the screening. The study protocol was approved by the ethics committee of the Bureau of Tainan County, and informed consent was obtained from all participants. Classification of aga The diagnosis of AGA was based on the pattern of hair loss of the participants. Assessment of the degree of hair loss was obtained by independent public health nurses trained by a dermatologist. To classify the degree of AGA for each subject, the classification by Norwood,2 a standard classification scheme with good test-retest reliability, was used. Some men with “female pattern hair loss” (noninvolvement of the frontal hairline) were assessed separately using the classification by Ludwig.9 Androgenetic alopecia among men with cosmetically significant male pattern baldness was defined as type III or greater (types III [including IIIv and IIIa], IV [including IVa], V [including Va], VI, and VII) according to the Norwood classification. The prevalence of more advanced degrees of alopecia characterized by only a remaining horseshoe fringe of hair (Norwood types V, Va, VI, and VII) was also estimated. The prevalence of AGA with female pattern hair loss was reported separately. Norwood type IV of AGA represents the starting grade of severe frontal AGA concurrent with vertex AGA. Therefore, we divided AGA into 2 categories (mild AGA [Norwood types I-III] and moderate or severe AGA [Norwood types IV-VII]) to assess its association with smoking and other potential risk factors. Smoking and other possible risk factors In addition to classification of AGA, we collected information on age at onset of AGA together with smoking and other possible risk factors using a face-to-face questionnaire interview. Smoking status (never, quit, or current); age at start of smoking; and quantity, duration, and frequency of smoking were collected using a structured questionnaire administered by trained staff members. In addition to smoking status, we categorized cigarette smoking into 4 groups (never, quit, current smoker of <20 cigarettes per day, or current smoker of ≥20 cigarettes per day) for comparison. Regarding anthropometric measures, body weight, body height, waist circumference, and hip circumference were measured. Body mass index was calculated as weight in kilograms divided by height in meters squared. Blood pressure was measured twice with at least a 5-minute interval between measurements. Hypertension was defined as a systolic blood pressure of 140 mm Hg or higher and a diastolic blood pressure of 90 mm Hg or higher or as taking antihypertensive medication. To collect biochemical markers, a venous blood sample was obtained after 12-hour fasting to check blood glucose, triglycerides, total cholesterol, and high-density lipoprotein cholesterol levels. Low-density lipoprotein cholesterol level was calculated in accord with the following formula: low-density lipoprotein cholesterol level = [(total cholesterol level − high-density lipoprotein cholesterol level − (triglyceride level/5)]. If the triglyceride level was greater than 400 mg/dL (to convert to millimoles per liter, multiply by 0.0113), the low-density lipoprotein cholesterol level was set as a missing value. Abnormal status for these biochemical markers included hypertriglyceridemia (triglyceride level, >160 mg/dL), hypercholesterolemia (total cholesterol level, >220 mg/dL [to convert to millimoles per liter, multiply by 0.0259]), low high-density lipoprotein cholesterol level (<42 mg/dL), high low-density lipoprotein cholesterol level (≥140 mg/dL), and high fasting glucose level (>110 mg/dL [to convert to millimoles per liter, multiply by 0.0555]). Dyslipidemia was defined as abnormal serum triglyceride, total cholesterol, high-density lipoprotein cholesterol, or low-density lipoprotein cholesterol level or as taking lipid-lowering drugs. In each subject, a questionnaire regarding diagnoses of chronic diseases such as hypertension, diabetes mellitus, cardiovascular disease, antihypertensive and lipid-lowering drug use, timing of growth spurt in puberty, socioeconomic factors, alcoholic beverage consumption, and betel nut chewing history was completed. The socioeconomic factors included age, occupation, and level of education. Age was defined as the age at study enrollment. Information about baldness among the first-degree, second-degree, and third-degree relatives was also collected using a structured questionnaire administered by trained staff members. The degree of hair loss was recalled by the study subjects. Statistical analysis The age-specific prevalence of AGA (Norwood types ≥III) was expressed as a percentage. A multivariate logistic regression model was used to assess the relationships of each possible risk factor such as smoking and degrees of relative relationships for family history of AGA (first-degree, second-degree, and third-degree relatives) to moderate or severe AGA (Norwood types ≥IV). A cumulative logits model was used to assess the incremental effect of age at onset of AGA (≤40 vs >40 years) on AGA severity (Norwood types I-VII) with ordinal properties. The proportional odds assumption was checked for the 7 AGA grades. In addition, the association between family history of AGA and age at onset of AGA was assessed. Because age and family history of AGA were 2 established factors for AGA, they were retained in the model in the multivariate analysis. All odds ratios (ORs) and their 95% confidence intervals (CIs) were computed using statistical software (SAS version 8; SAS Institute, Inc, Cary, North Carolina). Results Of 924 invited men, 740 participated in this survey. The mean ± SD age of participants was 65.2 ± 11.2 (age range, 40-91 years). Prevalence The age-specific number and percentage of different AGA types are given in Table 1Figure 1 shows that the age-specific prevalence of AGA (Norwood types III [including IIIv and IIIa], IV [including IVa], V [including Va], VI, and VII) increased with advancing age. The age-specific prevalence of severe hair loss (Norwood types V, Va, VI, and VII) was low. In Figure 2, the age-specific prevalence of AGA (Norwood types ≥III) is compared with the results of previous studies.2,7,10-15 Of 740 men examined, 48 (6.5%) were Norwood type A variants, higher than that reported in the study by Norwood (3%), and 13 (1.8%) manifested female pattern AGA. The prevalence did not increase with age for Norwood type A variants or for Ludwig grades L-I to L-III. For the groups aged 40 to 49, 50 to 59, 60 to 69, and 70 years or older, the age-specific prevalences of Norwood type A variants were 5.1%, 1.1%, 8.1%, and 7.4%, respectively, and those of female pattern AGA were 0.0%, 2.1%, 1.6%, and 2.4%, respectively (Table 1). Association In the univariate analysis, smokers were at increased risk of having moderate or severe AGA (Norwood types ≥IV) (OR, 1.61; 95% CI, 1.05-2.46). However, no statistically significant associations were found for intensity or duration of smoking (Table 2). There were statistically significant positive associations between moderate or severe AGA and smoking status (OR, 1.77; 95% CI, 1.14-2.76), current cigarette smoking of 20 cigarettes or more per day (OR, 2.34; 95% CI, 1.19-4.59), smoking intensity (OR, 1.78; 95% CI, 1.03-3.07), and dyslipidemia (OR, 1.47; 95% CI, 1.01-2.14) after adjusting for age and family history of AGA (Table 2). Smoking status and other statistically significant variables (age, family history of AGA, dyslipidemia, and betel nut chewing) were retained in the model and were adjusted for each other. A statistically significant association between smoking status and moderate or severe AGA (OR, 1.63; 95% CI, 1.00-2.65) still remained. Early-onset aga and aga grades The incremental effect of early-onset on severity of AGA was demonstrated by using the cumulative logits model. Findings suggest that earlier onset of AGA was associated with severe AGA. In those with early-onset AGA, AGA grades were statistically significantly worse (OR, 2.09; 95% CI, 1.14-3.85). Early-onset AGA was also related to family history of AGA (OR, 2.89; 95% CI, 1.42-5.88) using the logistic regression model. Risk for moderate or severe AGA increased with degree of relationship for family history (first degree [OR, 13.38; 95% CI, 4.80-37.27], second degree [OR, 6.33; 95% CI, 2.37-16.91], and third degree [OR, 5.32; 95% CI, 0.80-35.48]) (Table 3). The association between moderate or severe AGA and family history of AGA from paternal relatives was statistically significant (OR, 12.69; 95% CI, 4.65-34.60), whereas the corresponding association from maternal relatives was not (OR, 3.07; 95% CI, 0.30-31.32). The OR for the association between family history in both paternal and maternal relatives could not be estimated because there was no case with a positive family history from both paternal and maternal relatives among the subjects without AGA. Comment In our study, the age-specific prevalence of male pattern AGA was comparable with that among Korean men.12 Compared with the results of another Asian study15 conducted in Singapore, our prevalence rate of AGA was lower. However, the comparison should be interpreted with caution because the definition of cosmetically significant AGA by Tang et al15 was Norwood type I or greater, which was different from the definition (Norwood types ≥III) used in our study. This may also explain why the age-specific prevalence of AGA in their study was much higher than that in our study. The age-specific prevalence of AGA in another Asian study13 conducted in Thailand was also higher than that in our study. However, race/ethnicity in Taiwan is different from that in Thailand. The prevalence of AGA increased steadily with advancing age but was lower than that among persons of white race/ethnicity. An increase in risk of AGA with age reflects the natural progression of this condition. Advancing age is a risk factor for AGA in all men, irrespective of their family history. In addition, female pattern hair loss was observed in 13 of 740 men (1.8%) examined. Because AGA in some men may follow the female (Ludwig9) pattern, we suggest that the female pattern should be included in the classification of AGA in men. A positive association between smoking and AGA was demonstrated in our study. Our results were consistent with findings by Mosley and Gibbs5 using a cross-sectional survey in a general surgical outpatient clinic in the United Kingdom. Unlike our study that was based on the general population, the generalizability of their study may be limited because respondents came from a general surgical outpatient clinic. In addition, no confounders were considered, and the dose-response relationship to AGA with respect to amount and duration of smoking was not elucidated. Our positive results were not comparable with the findings from 2 studies.6,7 Severi et al7 reported a lower risk of AGA among current smokers (OR, 0.86; 95% CI, 0.54-1.38) and exsmokers (OR, 0.91; 95% CI, 0.65-1.29), although the results were not statistically significant. Nevertheless, details of smoking history were lacking, and the age at onset of AGA was unknown. In addition, different AGA classification categories instead of the Norwood classification were used. Therefore, it was difficult to compare their results with those of other studies. Matilainen et al6 found that the prevalence of current smoking among women with extensive hair loss (Ludwig grades ≥L-II to L-III) did not differ from those with normal or minimal hair loss (P =.92). However, the main objective of the study was not tailored to smoking history, and no definition of smoking was addressed. The mechanisms by which smoking causes hair loss may be multifactorial.1 First, cigarette smoking may be deleterious to the microvasculature of the dermal hair papilla. Second, smoke genotoxicants may do damage to DNA of the hair follicle. Third, smoking may lead to an imbalance in the follicular protease or antiprotease system. Smoking-induced oxidative stress may lead to the release of proinflammatory cytokines that, in turn, results in follicular microinflammation and fibrosis. Fourth, cigarette smoking may yield a relative hypoestrogenic state by inducing increased hydroxylation of estradiol and inhibition of aromatase.1 Although the genetic basis of AGA is well documented in the medical community and among the general population, there are few studies investigating the familial aggregation of AGA. Chumlea et al16 reported that the risk of AGA increased among men with a positive maternal grandfather history and even more with a history of hair loss in their father. In our study, moderate or severe AGA was statistically significantly associated with family history of AGA among the first-degree and second-degree relatives but not among the third-degree relatives after adjusting for age. The highest OR was associated with a family history among first-degree relatives. In addition, family history of AGA among paternal relatives was statistically significantly predictive of moderate or severe AGA after adjusting for age. Our findings do not support an association between moderate or severe AGA and family history of AGA among maternal relatives. The effect of family history among both paternal and maternal older relatives could not be assessed because no case was found among subjects without AGA. Moreover, we found that family history of AGA is statistically significantly associated with the risk of early-onset AGA. This implies that those with a family history of AGA may have a higher risk of early-onset AGA and a higher risk of developing severe AGA. Most important, early-onset AGA (age at onset ≤40 years) showed a statistically significant dose-dependent association with AGA grade after adjusting for age and family history. From the clinical point of view, this suggests that patients with early-onset AGA should receive early advice to prevent the deterioration of AGA. In conclusion, our study showed that the prevalence of AGA among men in Taiwan was lower than that among persons of white race/ethnicity. The severity of AGA was also milder in Taiwan. We suggest that a female pattern should be included in the classification because AGA in some men may follow this Ludwig pattern. We also confirmed that smoking status, current amount of cigarette smoking, and smoking intensity play important roles in the development of moderate or severe AGA. Correspondence: Tony Hsiu-Hsi Chen, DDS, PhD, Division of Biostatistics, Institute of Preventive Medicine, College of Public Health, National Taiwan University, Room 521, Fifth Floor, 17 Suchow Rd, Taipei 100, Taiwan (chenlin@ntu.edu.tw). Financial Disclosure: None reported. Back to top Article Information Accepted for Publication: May 3, 2007. Author Contributions: Drs Su and Chen had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Chen. Acquisition of data: Su. Analysis and interpretation of data: Chen and Su. Drafting of the manuscript: Su. Critical revision of the manuscript for important intellectual content: Chen. Statistical analysis: Chen. Administrative, technical, and material support: Chen. Study supervision: Chen. Additional Contributions: Yao-Der Chen, MS, and other colleagues helped with data collection, and Min-Shu Hsu, BSc, performed data management. References 1. Trüeb RM Molecular mechanisms of androgenetic alopecia. Exp Gerontol 2002;37 (8-9) 981- 990PubMedGoogle ScholarCrossref 2. Norwood OT Male pattern baldness: classification and incidence. South Med J 1975;68 (11) 1359- 1365PubMedGoogle ScholarCrossref 3. Hoffmann R Male androgenetic alopecia. Clin Exp Dermatol 2002;27 (5) 373- 382PubMedGoogle ScholarCrossref 4. Ellis JASinclair RHarrap SB Androgenetic alopecia: pathogenesis and potential for therapy. Expert Rev Mol Med 2002;20021- 11PubMedGoogle Scholar 5. Mosley JGGibbs AC Premature grey hair and hair loss among smokers: a new opportunity for health education? BMJ 1996;313 (7072) 1616PubMedGoogle ScholarCrossref 6. Matilainen VLaakso MHirsso PKoskela PRajala UKeinanen-Kiukaanniemi S Hair loss, insulin resistance, and heredity in middle-aged women: a population-based study. J Cardiovasc Risk 2003;10 (3) 227- 231PubMedGoogle ScholarCrossref 7. Severi GSinclair RHopper JL et al. Androgenetic alopecia in men aged 40-69 years: prevalence and risk factors. Br J Dermatol 2003;149 (6) 1207- 1213PubMedGoogle ScholarCrossref 8. Chen THChiu YHLuh DL et al. Taiwan Community-Based Integrated Screening Group, Community-based multiple screening model: design, implementation, and analysis of 42,387 participants. Cancer 2004;100 (8) 1734- 1743PubMedGoogle ScholarCrossref 9. Ludwig E Classification of the types of androgenetic alopecia (common baldness) occurring in the female sex. Br J Dermatol 1977;97 (3) 247- 254PubMedGoogle ScholarCrossref 10. DeMuro-Mercon CRhodes TGirman CJVatten L Male-pattern hair loss in Norwegian men: a community-based study. Dermatology 2000;200 (3) 219- 222PubMedGoogle ScholarCrossref 11. Hamilton JB Patterned loss of hair in man: types and incidence. Ann N Y Acad Sci 1951;53 (3) 708- 728PubMedGoogle ScholarCrossref 12. Paik JHYoon JBSim WYKim BSKim NI The prevalence and types of androgenetic alopecia in Korean men and women. Br J Dermatol 2001;145 (1) 95- 99PubMedGoogle ScholarCrossref 13. Pathomvanich DPongratananukul SThienthaworn PManoshai S A random study of Asian male androgenetic alopecia in Bangkok, Thailand. Dermatol Surg 2002;28 (9) 804- 807PubMedGoogle Scholar 14. Rhodes TGirman CJSavin RC et al. Prevalence of male pattern hair loss in 18-49 year old men. Dermatol Surg 1998;24 (12) 1330- 1332PubMedGoogle ScholarCrossref 15. Tang PHChia HPCheong LLKoh D A community study of male androgenetic alopecia in Bishan, Singapore. Singapore Med J 2000;41 (5) 202- 205PubMedGoogle Scholar 16. Chumlea WCRhodes TGirman CJ et al. Family history and risk of hair loss. Dermatology 2004;209 (1) 33- 39PubMedGoogle ScholarCrossref

Journal

Archives of DermatologyAmerican Medical Association

Published: Nov 1, 2007

There are no references for this article.