Context: Perimenopause is a time of transition for women at midlife. Women want to know whether they are starting this change and physicians need to know the accuracy of a clinical examination in identifying perimenopausal women. These women should be counseled about alleviating climacteric symptoms, using contraception, and preventing diseases such as osteoporosis.
Objective: To systematically review the accuracy of self-assessment, symptoms, signs, and laboratory tests in diagnosing women in perimenopause.
Data sources: English-language articles that presented data relevant to diagnosis of perimenopause were identified in a MEDLINE search from 1966 to 2001. References of these articles and other publications also were reviewed.
Study selection: Cross-sectional or longitudinal studies of women aged 40 years or older that used the definition of perimenopause as 3 to 11 months of amenorrhea or irregular periods, included a premenopausal control group, and reported a clinical examination finding. Of 1246 articles identified, 16 studies were included in the analysis.
Data extraction: Two authors independently reviewed articles for quality (L.A.B. and C.M.S.). Discrepancies were resolved by a third author (K.N.).
Data synthesis: The prior probability of perimenopause is directly related to a woman's age. After considering age, the following yielded the greatest positive likelihood ratios (LRs+): self assessment of going through the transition (LR+ range, 1.53-2.13), symptoms of hot flashes (LR+ range, 2.15-4.06), night sweats (LR+ 1.90; 95% confidence interval [CI], 1.63-2.21), vaginal dryness (LR+ range, 1.48-3.79), high follicle-stimulating hormone levels (LR+ 3.06; 95% CI, 2.06-4.54), and low inhibin B levels (LR+ 2.05; 95% CI, 0.96-4.39). Self-assessment of perimenopausal status had the smallest negative LR (range, 0.18-0.36).
Conclusions: No one symptom or test is accurate enough by itself to rule in or rule out perimenopause. Clinicians should diagnose perimenopause based on menstrual history and age without relying on laboratory test results.