- Email: [email protected]
Severity of mastalgia in relation to milk duct dilatation
Severity of Mastalgia in Relation to Milk Duct Dilatation Friedolf Peters, PhD, Petra Diemer, MD, Olaf Mecks, and Lu¨tje J. Behnken, PhD OBJECTIVE: The etiology of mastalgia is poorly understood. Histology cannot detect any distinct characteristics. This investigation therefore applied ultrasonography to mastalgia patients to investigate morphological structures and to obtain further insights into the pathophysiology of mastalgia. The aim of the study was to analyze the significance of milk duct dilatation in patients with mastalgia. METHODS: A total of 335 premenopausal women participated in a genital and breast cancer screening program. Of these, 123 women were asymptomatic and 212 complained of breast pain (136 had cyclical mastalgia; 76 reported the noncyclical type). The width of milk ducts was measured by ultrasonography and correlated with the intensity of breast pain. The site of the pain was correlated with duct dilatation. Statistical analysis was performed using the t test, the 2 test, analysis of variance, and Pearson correlation. RESULTS: In asymptomatic women, the maximum mean width of the milk ducts was 1.8 ⴞ 0.84 mm, in cyclical mastalgia 2.34 ⴞ 1.10 mm, and 3.89 ⴞ 1.26 mm in noncyclical mastalgia (P < .001). The intensity of pain showed a significant, positive correlation with the width of the milk ducts (r ⴝ .5008, P < .001). Noncyclical mastalgia patients located the pain at the site where dilated ducts were detected ultrasonographically (P < .001). CONCLUSION: The results of this study demonstrate that duct ectasia is a major factor in mastalgia. The degree of duct dilatation correlates positively with the intensity of breast pain. In the noncyclical type of pain, there is a positive correlation between the site of duct dilatation and the site of pain. (Obstet Gynecol 2003;101:54 – 60. © 2003 by The American College of Obstetricians and Gynecologists.) From the Department of Obstetrics and Gynecology, St. Hildegardis Hospital, Academic Teaching Hospital, Mainz; and Bioscientia Institute of Laboratory Chemistry, Ingelheim, Germany. These results are part of the doctoral thesis of Petra Diemer. The authors are indebted to the mathematician B. Nafe for statistical analysis and to Martin Hahn, MD, for his critical advice. This study is dedicated to Professor Dr. Klaus-Dieter Schulz, President of the World Society of Breast Health, in appreciation of his indefatigable efforts to improve women’s health care.
54
Public awareness of breast cancer causes concern in every patient with breast symptoms. The most common complaints are a lump, nipple discharge, inflammatory signs, and breast pain. The nature of the tumor can be established by an image-guided biopsy,1 and guidelines for evaluating nipple discharge have been published.2 We have recently described the coincidence of nonpuerperal mastitis and breast cancer and the procedure to investigate this condition.3 Breast pain is of considerable importance within the health care system. About 8 –10% of all women seek medical advice for breast pain at least once during their life after menarche. Pain may be a presenting symptom of a small breast cancer.4 Therefore, breast cancer has to be ruled out using the established procedures in this most common breast symptom; however, knowledge of the etiology of breast pain is incomplete, and medical means of resolving the disorder are rather unsatisfactory. The cause of cyclical mastalgia is unknown. Histopathology also cannot detect any distinctive characteristics.5,6 From a clinical point of view, mastalgia patients are divided into subgroups of which the cyclically pronounced, the noncyclical pattern, and the Tietze syndrome are the most frequent types.7 It is believed that cyclical mastalgia is hormonally based7,8 or mediated by fat intake, especially of saturated fatty acids.9 We characterized mastalgia endocrinologically in terms of prolactin secretion8 and thyroid hormones.10 Moreover, mastalgia responds to hormone treatment 11,12 and fat restriction13 or administration of essential fatty acids.12 Noncyclical mastalgia has been associated with duct ectasia on the basis of histological findings or mammograms7; however, milk ducts can be visualized by mammography only in cases of tissue reaction, such as periductal mastitis, fibrosis, or intraductal calcification or fibrosis.14 Moreover, the feasibility of identifying any consistent relationship between mammographic patterns and histology seems questionable.15 Furthermore, mammography cannot exactly localize milk ducts without further equipment, such as stereotaxy or wires. Technological advances of ultrasonography, such as color
VOL. 101, NO. 1, JANUARY 2003 © 2003 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
0029-7844/03/$30.00 PII S0029-7844(02)02386-4
Table 1. Clinical Data for 335 Premenopausal Women
Age (y) (mean ⫾ SD) Parity (mean) Maximum dilatation (mm) (mean ⫾ SD) Intensity of pain† (mean ⫾ SD) Dilated ducts ⫽ site of pain [n (%)]
Asymptomatic (n ⫽ 123)
Cyclical mastalgia (n ⫽ 136)
Noncyclical mastalgia (n ⫽ 76)
33.50 ⫾ 6.00 1.35 1.80 ⫾ 0.84
32.85 ⫾ 8.70 1.38 2.34 ⫾ 1.10 34.7 ⫾ 17.2 24 (17.7)
34.32 ⫾ 6.69 1.30 3.89 ⫾ 1.26* 53.0 ⫾ 19.3* 63 (82.9)*
*P ⬍ .001 t test. † Pain was patient-assessed using a 100-mm analogue scale.
Doppler imaging, high-frequency ultrasound, and contrast media, are increasing the specificity in evaluating benign conditions of the breast. Ultrasound has been deemed appropriate to obtain further insights into breast structures with apparently benign conditions. First attempts to apply ultrasonography to patients with mastalgia used low-frequency probes and could only detect cysts.16 The milk ducts of the breast have been neglected by clinicians in the evaluation of breast symptoms and disorders for a long time. It was believed that the milk ducts remained asymptomatic unless they had been distended during lactation or inflammation (duct ectasia with periductal mastitis and breast infection).7 In most women, it is possible to image the milk ducts with modern sonographic techniques even if they are not distended. This study therefore investigated the ultrasonographic appearance of the milk ducts in otherwise healthy women to obtain further insights into the nature of mastalgia and to correlate the width of the ducts with the severity of mastalgia. MATERIALS AND METHODS A total of 361 premenopausal, otherwise healthy women participated in a genital and mammary cancer screening program between 1998 and 1999 (mean age 33.43 ⫾ 7.23 [standard deviation, SD], range 17– 47 years). These women visited the outpatient clinic once a year on a regular basis or when a special symptom required examination. Breast cancer was ruled out by clinical examination and by mammography in patients older than 32 years. Ten patients who had undergone prior breast surgery or who needed further evalution of a breast mass detected by palpation, mammography, or ultrasound were excluded from the study, as were 12 patients who needed treatment for a tumor of the adnexa or for cervical intraepithelial neoplasia. Four patients were excluded because of concurrent use of antidepressants or analgesics. The remaining 335 patients participated in the study. All were white and of German origin. They were not receiving any regular medication or
VOL. 101, NO. 1, JANUARY 2003
special diet. Parity and use of hormonal contraception was documented. Each woman was given an ultrasound scan to rule out a nonpalpable tumor or a tumor not seen in the mammography. Special attention was focused on the milk ducts. All four quadrants of the breast and the retroareolar region were imaged systematically, and the presence and the width of the ducts were documented. After this examination, each woman was given a pain chart and a linear visual analogue scale (0 –100 mm, 0 ⫽ no pain, 100 ⫽ worst pain) to identify any breast pain and its intensity.17 Pain was recorded every day throughout the entire cycle. After judging the pain as cyclic or noncyclic according to Preece and colleagues,7 the day of the most intensive pain was assessed for the study. The patients were asked about recent typical symptoms of duct ectasia and periductal mastitis, such as nipple discharge, nonlactational infection, nipple retraction, and breast lump, and about smoking habits. Patients with breast pain were given another ultrasound scan during the period of the worst pain to correlate the site of the pain with the site of the dilated ducts. The sonographic picture was analyzed for a putative tissue edema. Homogeneous echogenicity of the parenchyma differing from fibrous structures in conjunction with the patient’s report of breast swelling was interpreted as tissue edema. Tissue edema could not be quantified. For these ultrasonographic scans, an AI 3200 (Dornier Instruments, Munich, Germany) was used with a linear array 7.5-MHz probe. This instrument can focus on different depths of the breast. Large breasts and small breasts could be examined with the same precision. The method successfully imaged the milk ducts in different parts of the breast. In some cases, a special maneuver or conformational change to the breast was necessary, as described by Harper and Kelly-Fry,18 to enable the ducts to be visualized. To clearly distinguish the milk ducts from the blood vessels, color Doppler sonography was applied. The lower threshold of this ultrasound technique in detecting a milk duct was a dilatation of more than 0.4 mm. In each patient, the milk ducts were identified, and their maximum width was evaluated. The
Peters et al
Mastalgia and Milk Duct Dilatation
55
Table 2. Patient Histories and Details of Study Findings Cyclical Noncyclical Asymptomatic mastalgia mastalgia (n ⫽ 123) (n ⫽ 136) (n ⫽ 76) Nipple discharge Nipple retraction Breast infection Lump Supposed tissue edema Nicotine addiction
2 (1.6)
5 (3.7)
2 (1.6)
9 (6.6) 63 (46.3)
13 (17.1) 1 (1.3) 2 (2.6) 3 (3.9) 3 (3.9)
14 (11.4)
20 (14.7)
37 (48.7)
Values are n (%).
ultrasound scans were exclusively done by one person (PD). The intraobserver variability of the width of the milk ducts was (mean ⫾ SD) 4.8% ⫾ 1.8% (n ⫽ 40). A prior study investigated whether there was any significant difference in the diameter of the milk ducts in the course of the menstrual cycle. Forty patients of the three groups with measurable milk ducts were randomly selected and given two separate ultrasound scans on days 7–10 and on days 24 –26 of the cycle. Statistical analysis was performed with the t test, the 2 test, correlation coefficient according to Pearson, and an analysis of variance. RESULTS In the entire study group, the width of the milk ducts measured ranged from 0.4 mm to 8.0 mm. Milk ducts
were undetectable in 21 women (⬍0.4 mm). One hundred twenty-three patients reported no symptoms in their breasts; 212 patients complained of pain. From a total of 212 mastalgia patients, 136 (64.2%) had cyclically pronounced and 76 (35.8%) had noncyclical mastalgia (Table 1). The mean age of the three groups (asymptomatic: 33.5 ⫾ 6.00 years, cyclically pronounced: 32.85 ⫾ 8.70 years, noncyclical: 34.32 ⫾ 6.69 years) did not vary statistically (analysis of variance). Details regarding clinical symptoms and nicotine addiction are listed in Table 2. Dilated milk ducts were found throughout all the quadrants of the breast, predominantly in the retroareolar region (Figure 1). In the 40 patients of the three groups with measurable milk ducts who were investigated during the first and the second part of the cycle, the width of the milk ducts did not vary significantly: follicular phase: 2.15 ⫾ 1.8 mm, luteal phase: 2.34 ⫾ 1.6 mm, P ⬎ .05 (t test for paired data). Therefore, the day of the cycle was not mentioned in this study. In asymptomatic women, milk duct widths from 0.6 mm up to 4.4 mm were measured. The mean maximum width was 1.8 ⫾ 0.84 mm in asymptomatic women (Table 1), 2.34 ⫾ 1.10 mm in cyclically pronounced (range 0.5–5.2 mm), and 3.89 ⫾ 1.26 mm (range 1.3– 8.0 mm) in noncyclical mastalgia (t210 ⫽ 9.30, P ⬍ .001). A tissue edema was diagnosed in 1.6% of asymptomatic women, in 46.3% of cyclical, and in 3.9% of noncyclical mastalgia patients (Table 2). Representative depictions
Figure 1. Allocation of milk ducts (⬎0.4 mm) to the particular quadrant and the areola of the breast in 64 patients. The circle in the top right corner of the figure describes the quadrants of the breast: 1 and 4 ⫽ outer quadrants, A ⫽ areolar region. Peters. Mastalgia and Milk Duct Dilatation. Obstet Gynecol 2003.
56
Peters et al
Mastalgia and Milk Duct Dilatation
OBSTETRICS & GYNECOLOGY
and 24 (17.7%) of the cyclically pronounced mastalgia patients pinpointed the pain at the site where the dilated ducts had been detected ultrasonographically (2 ⫽ 740.69, P ⬍ .001).
Figure 2. Representative ultrasonography of the breast of the three groups. a) Characteristic texture of a normal breast (asymptomatic) with clearly visible Cooper ligaments as well as breast and fat tissue between them. b) A narrow milk duct detectable behind the areola. c) Breast tissue of the upper outer quadrant of a patient with cyclical mastalgia. The normal structure (a) is covered with a “veil,” suggesting the clinically evident breast edema. d) Milk ducts in patients with cyclically pronounced mastalgia were found to be wider than in asymptomatic subjects. e) Diffused dilated milk ducts in a patient with noncyclical mastalgia (upper outer quadrant of the breast). f) A corresponding milk duct that is clearly dilated. Peters. Mastalgia and Milk Duct Dilatation. Obstet Gynecol 2003.
of the asymptomatic breast, premenstrual mastalgia, and noncyclical mastalgia with the corresponding milk ducts are presented in Figure 2. The severity of the pain (visual linear analogue scale) in noncyclical mastalgia (53.0 ⫾ 19.3 mm) significantly exceeded that of cyclically pronounced mastalgia (34.7 ⫾ 17.2 mm, t210 ⫽ 7.12, P ⬍ .001). The intensity of pain showed a significant, positive correlation with the width of the milk ducts (Figure 3, r ⫽ .5008, P ⬍ .001). Sixty-three (82.9%) of the noncyclical
VOL. 101, NO. 1, JANUARY 2003
DISCUSSION The data presented here indicate that dilatation of the milk ducts from a certain degree onward does indeed produce breast pain in nonlactating women. Regarding symptomatic mastalgia, duct ectasia is a major factor in this disorder, nearly exclusively so in noncyclical mastalgia, and it affects at least one third of the patients with cyclical mastalgia. Cyclical and noncyclical pain were evaluated together in the correlation analysis (Figure 1). A more correct analysis might be made by dividing the cases into breast pain groups; however, the aim of this study was to investigate a relationship between breast pain and the degree of dilatation of the ducts. The numerical distribution of women in the three groups neither reflects the actual incidence of mastalgia within the population nor the ratio between the two types of breast pain. The women were recruited from a cancer screening program, therefore patients with breast pain who were worried about having breast cancer are overrepresented. Furthermore, the asymptomatic group does not represent the normal population as control subjects. Therefore, bias is possible. There is a good correlation between the maximum width of the milk ducts and the degree of the pain documented. Patients with wider milk ducts have more pain than those with narrower ducts. The majority of our cases involved the terminal ducts and the subareolar region, sometimes only over a short distance. Moreover, the localized pain can be attributed to an ultrasonographically identified dilated duct at this site. In a noncyclical mastalgia group of patients, clinical examination by Griffith and colleagues found a palpable ectatic duct behind the nipple, with green- or yellow-stained nipple discharge or with a trigger zone pain (persistent and well localized).19 This correlates with our finding that ultrasonography reveals a dilated duct at the site of pain. Until now, the cause of mastalgia has not been clarified conclusively. In cyclical mastalgia, elevated prolactin response to thyrotropin-releasing hormone (TRH), and thus a higher daily prolactin production, has been documented.8 Bromocriptine as a potent prolactin-lowering drug is established in the treatment of cyclical mastalgia.12 A possible mechanism of action of prolactin may be fluid and electrolyte absorption. Although the effects of prolactin on osmoregulation in mammals are less clear, it has been shown that prolactin stimulates the retention of fluid and electrolytes in the intestine and
Peters et al
Mastalgia and Milk Duct Dilatation
57
Figure 3. Correlation between the width of the milk ducts and the intensity of breast pain (n ⫽ 168, r ⫽ 0.5008, P ⬍ .001). max ⫽ maximum. Peters. Mastalgia and Milk Duct Dilatation. Obstet Gynecol 2003.
kidney in rats.20,21 Moreover, prolactin may also play a role in osmoregulation in amniotic fluid by decreasing the diffusion of water from the fetal side to the maternal side of the membrane.22 Prolactin-related fluid retention within the breast may provide a possible explanation for premenstrual mastalgia. The increase of the breast volume during the premenstrual period is a well-known phenomenon.23 The tissue edema could be inferred from ultrasonography in some cases in our study, although it is difficult to quantify this phenomenon. Tissue edema may be responsible for a tension-mediated pain7,12,23 and may result in a compression of the milk ducts that are narrower than those in noncyclical mastalgia. Because breast secretion is thought to be controlled by prolactin,24 one would expect the wider milk ducts in patients with noncyclical mastalgia to be correlated with elevated prolactin levels. However, an earlier study by Kumar and colleagues,25 as well as a conjoined study by our group, could detect only normal basal and TRHstimulated prolactin values (manuscript in preparation). The response rate to prolactin-lowering treatment with bromocriptine is also less in patients with noncyclical mastalgia compared with patients with the cyclical type.12 Thus, cyclical and noncyclical mastalgia are probably two distinct entities. The reason why milk ducts are more distended in noncyclical mastalgia than in the cyclical type may at least be related to the tissue edema in the latter group. Because milk ducts also re-
58
Peters et al
Mastalgia and Milk Duct Dilatation
gress in width in postmenopausal women, attention should also be focused on estrogens. Further studies are needed to elucidate possible connections. Whatever the cause of distended milk ducts is, their manifestation is correlated with pain, as documented in our study. Milk ducts are known to be sensitive to pain. Every breastfeeding mother describes discomfort or pain of the filled breast before letting the child suckle. Breast engorgement during lactation can also be extremely painful. Therefore, the dilated milk ducts described by this study are most likely to be the origin of the pain, which is mediated by sensory nerve endings accompanying the blood vessels.26 Normal measurements for the diameter of a milk duct are not clearly defined. Regarding the histological criteria of duct ectasia with periductal mastitis, most authors describe milk ducts with a diameter of roughly 3 mm as fulfilling the definition.27 Based on our data, a cutoff value of 3 mm (eg, as a criterion for defining a dilated duct) is not clinically feasible in the same way. Most asymptomatic women have duct widths below this value, and most symptomatic patients have widths above it. Statistically, a threshold of maximum ductal dilatation that predicts breast pain cannot be evaluated on the basis of our data. The term duct ectasia is frequently associated with periductal mastitis. Two theories for its development have been formulated. The first proposes that milk ducts dilate with age and that stagnant secretion in these di-
OBSTETRICS & GYNECOLOGY
lated ducts leaks into the surrounding tissue to cause periductal mastitis.28 The second theory specifies that periductal inflammation predominates in younger women, whereas duct dilatation increases in frequency with advancing age.29,30 Periductal mastitis is frequently combined with clinical episodes of periductal mastitis and abscess formation.30 This especially affects women between the ages of 18 and 45.3,30 In our study, however, only two women reported a recent clinical mastitis, whereas typical symptoms of duct ectasia, such as nipple discharge and nicotine addiction,30 were mainly observed in patients with noncyclical mastalgia, in whom the greatest proportion of duct ectasia was seen. Breast biopsy was not indicated in the patients of this study. Consequently, we cannot report any data on periductal mastitis in noncyclical mastalgia. Because this clinical study found dilated ducts within an age group that is not typical for histologically proven duct ectasia,30 it seems possible that duct ectasia is a symptomatic variation of this condition that develops earlier in life. The reason milk ducts are dilated in patients with mastalgia is unknown; however, the study presented here provides a reasonable explanation for the general morphology underlying this disease. In conclusion, these results demonstrate for the first time that duct ectasia is a major factor in determining the severity of mastalgia. The site of pain is positively correlated with the site of duct dilatation in a considerable number of patients. REFERENCES 1. Venta LA. Image-guided biopsy of nonpalpable breast lesions. In: Harriss JR, Lippman ME, Morrow M, Osborne CK, eds. Diseases of the breast, 2nd ed. Philadelphia: Lippincott, Williams & Wilkins, 2000:149–64. 2. King TA, Carter KM, Bolton JS, Fuhrmann GM. A simple approach to nipple discharge. Arch Surg 2000;69:960–5. 3. Peters F, Kielich A, Pahnke V. Coincidence of nonpuerperal mastitis and breast cancer. Eur J Obstet Gynecol Reprod Biol 2002. In press. 4. Preece PE, Baum M, Mansel RE, Webster DJ, Fortt RW, Gravelle ICH, et al. Importance of mastalgia in operable breast cancer. BMJ 1982;248:1299–300. 5. Watt-Boolsen S, Emus H, Junge J. Fibrocystic disease and mastalgia. Dan Med Bull 1982;29:252–4. 6. Jorgensen J, Watt-Boolsen S. Cyclical mastalgia and breast pathology. Acta Chir Scand 1985;151:319–21. 7. Preece PE, Mansel RE, Bolten PM, Hughes LE, Baum M, Gravelle IH. Clinical syndromes of mastalgia. Lancet 1976;2:670–3. 8. Peters F, Schuth W, Scheurich B, Breckwoldt M. Serum prolactin levels in patients with fibrocystic breast disease. Obstet Gynecol 1984;64:381–5.
VOL. 101, NO. 1, JANUARY 2003
9. Goodwin PJ, Miller A, Del Giudice E, Singer W, Conelly P, Ritchie JWK. Elevated high-density lipoprotein cholesterol and dietary fat intake in women with cyclical mastopathy. Am J Obstet Gynecol 1998;179:430–7. 10. Peters F, Pickardt CR, Breckwold M. Thyroid hormones in benign breast disease: Normalization of exaggerated prolactin responsiveness to thyrotropin-releasing hormone. Cancer 1985;65:102–5. 11. Peters F. Multicentre study of gestrinone in cyclical breast pain. Lancet 1996;339:205–8. 12. Pye JK, Mansel RE, Hughes LE. Clinical experience of drug treatment for mastalgia. Lancet 1985;2:373–6. 13. Boyd NF, McGuire V, Shannon P, Cousins M, Kriukov V, Mahoney L, et al. Effect of a low-fat high-carbohydrate diet on symptoms of cyclical mastopathy. Lancet 1988;2: 128–32. 14. Wellings SR, Wolfe JN. Correlative studies of the histological and radiographic appearance of the breast parenchyma. Radiology 1978;129:299–306. 15. Moskowitz M, Gartside P, McLaughlin C. Mammographic patterns as markers for high risk benign breast disease and incident cancer. Radiology 1980;134:293–5. 16. Ayers J, Gidwani G. The “luteal breast” and hormonal and sonographic investigation of benign breast in patients with cyclical mastalgia. Fertil Steril 1983;408:779–84. 17. Tawaf-Mutamen H, Ader DN, Browne MW, Shriver CD. Clinical evaluation of mastalgia. Arch Surg 1998;133: 211–3. 18. Harper P, Kelly-Fry E. Ultrasound visualization of the breast in symptomatic patients. Radiology 1980;137: 465–9. 19. Griffith CDM, Dowle CS, Hinton CP, Blamey RW. The breast pain clinic: A rational approach to classification and treatment of breast pain. Postgrad Med J 1987;63:547–9. 20. Halloran BP, DeLuca HF. Calcium transport in small intestine during pregnancy and lactation. Am J Physiol 1992;239:64–8. 21. Stier CT, Cowden EA, Friesen HG. Prolactin and the rat kidney: A clearance and micropuncture study. Endocrinology 1984;115:362–7. 22. Tyson JE. The evolutionary role of prolactin in mammalian osmoregulation: Effects on fetoplacental hydromineral transport. Semin Perinatol 1982;6:216–28. ¨ ber Vera¨nderungen des Brustvolumens im 23. Do¨ring GK. U Cyclus. Arch Gynecol 1953;184:51–8. 24. Rose DP, Berke B, Cohen LA, Lahti H. A comparison of serum and breast duct fluid-immunoassayable prolaction and growth hormone with bioassayable lactogenic hormones in healthy women and patients with cystic breast disease. Cancer 1987;60:2761–5. 25. Kumar S, Mansel RE, Hughes LE, Woodhead JS, Edwards CA, Scanlon MF, et al. Prolactin response to thyrotropin-releasing hormone stimulation and dopami-
Peters et al
Mastalgia and Milk Duct Dilatation
59
26. 27.
28. 29.
60
nergic inhibition in benign breast disease. Cancer 1984; 153:1311–5. Vorherr H. The breast: Morphology, physiology and lactation. New York: Academic Press, 1974. Tedeschi L, Ahari S, Byrne J. Involutional mammary duct ectasia and periductal mastitis. Am J Surg 1963;106: 517–21. Haagensen CD. Mammary-duct ectasia: A disease which may simulate carcinoma. Cancer 1951;4:749–61. Bonser GM, Dossett JA, Jull JW. Human and experimental breast cancer. London: Pitman Medical, 1961:237–49.
Peters et al
Mastalgia and Milk Duct Dilatation
30. Dixon JM, Ravi Sekar O, Chetty U, Anderson TJ. Periductal mastitis and duct ectasia: Different conditions with different aetiologies. Br J Surg 1996;83:820. Address reprint requests to: Friedolf Peters, PhD, St. Hildegardis Hospital, Academic Teaching Hospital, Department of Obstetrics and Gynecology, D-55131 Mainz, Germany; E-mail: [email protected]. Received March 20, 2002. Received in revised form July 4, 2002. Accepted August 1, 2002.
OBSTETRICS & GYNECOLOGY
54
Public awareness of breast cancer causes concern in every patient with breast symptoms. The most common complaints are a lump, nipple discharge, inflammatory signs, and breast pain. The nature of the tumor can be established by an image-guided biopsy,1 and guidelines for evaluating nipple discharge have been published.2 We have recently described the coincidence of nonpuerperal mastitis and breast cancer and the procedure to investigate this condition.3 Breast pain is of considerable importance within the health care system. About 8 –10% of all women seek medical advice for breast pain at least once during their life after menarche. Pain may be a presenting symptom of a small breast cancer.4 Therefore, breast cancer has to be ruled out using the established procedures in this most common breast symptom; however, knowledge of the etiology of breast pain is incomplete, and medical means of resolving the disorder are rather unsatisfactory. The cause of cyclical mastalgia is unknown. Histopathology also cannot detect any distinctive characteristics.5,6 From a clinical point of view, mastalgia patients are divided into subgroups of which the cyclically pronounced, the noncyclical pattern, and the Tietze syndrome are the most frequent types.7 It is believed that cyclical mastalgia is hormonally based7,8 or mediated by fat intake, especially of saturated fatty acids.9 We characterized mastalgia endocrinologically in terms of prolactin secretion8 and thyroid hormones.10 Moreover, mastalgia responds to hormone treatment 11,12 and fat restriction13 or administration of essential fatty acids.12 Noncyclical mastalgia has been associated with duct ectasia on the basis of histological findings or mammograms7; however, milk ducts can be visualized by mammography only in cases of tissue reaction, such as periductal mastitis, fibrosis, or intraductal calcification or fibrosis.14 Moreover, the feasibility of identifying any consistent relationship between mammographic patterns and histology seems questionable.15 Furthermore, mammography cannot exactly localize milk ducts without further equipment, such as stereotaxy or wires. Technological advances of ultrasonography, such as color
VOL. 101, NO. 1, JANUARY 2003 © 2003 by The American College of Obstetricians and Gynecologists. Published by Elsevier Science Inc.
0029-7844/03/$30.00 PII S0029-7844(02)02386-4
Table 1. Clinical Data for 335 Premenopausal Women
Age (y) (mean ⫾ SD) Parity (mean) Maximum dilatation (mm) (mean ⫾ SD) Intensity of pain† (mean ⫾ SD) Dilated ducts ⫽ site of pain [n (%)]
Asymptomatic (n ⫽ 123)
Cyclical mastalgia (n ⫽ 136)
Noncyclical mastalgia (n ⫽ 76)
33.50 ⫾ 6.00 1.35 1.80 ⫾ 0.84
32.85 ⫾ 8.70 1.38 2.34 ⫾ 1.10 34.7 ⫾ 17.2 24 (17.7)
34.32 ⫾ 6.69 1.30 3.89 ⫾ 1.26* 53.0 ⫾ 19.3* 63 (82.9)*
*P ⬍ .001 t test. † Pain was patient-assessed using a 100-mm analogue scale.
Doppler imaging, high-frequency ultrasound, and contrast media, are increasing the specificity in evaluating benign conditions of the breast. Ultrasound has been deemed appropriate to obtain further insights into breast structures with apparently benign conditions. First attempts to apply ultrasonography to patients with mastalgia used low-frequency probes and could only detect cysts.16 The milk ducts of the breast have been neglected by clinicians in the evaluation of breast symptoms and disorders for a long time. It was believed that the milk ducts remained asymptomatic unless they had been distended during lactation or inflammation (duct ectasia with periductal mastitis and breast infection).7 In most women, it is possible to image the milk ducts with modern sonographic techniques even if they are not distended. This study therefore investigated the ultrasonographic appearance of the milk ducts in otherwise healthy women to obtain further insights into the nature of mastalgia and to correlate the width of the ducts with the severity of mastalgia. MATERIALS AND METHODS A total of 361 premenopausal, otherwise healthy women participated in a genital and mammary cancer screening program between 1998 and 1999 (mean age 33.43 ⫾ 7.23 [standard deviation, SD], range 17– 47 years). These women visited the outpatient clinic once a year on a regular basis or when a special symptom required examination. Breast cancer was ruled out by clinical examination and by mammography in patients older than 32 years. Ten patients who had undergone prior breast surgery or who needed further evalution of a breast mass detected by palpation, mammography, or ultrasound were excluded from the study, as were 12 patients who needed treatment for a tumor of the adnexa or for cervical intraepithelial neoplasia. Four patients were excluded because of concurrent use of antidepressants or analgesics. The remaining 335 patients participated in the study. All were white and of German origin. They were not receiving any regular medication or
VOL. 101, NO. 1, JANUARY 2003
special diet. Parity and use of hormonal contraception was documented. Each woman was given an ultrasound scan to rule out a nonpalpable tumor or a tumor not seen in the mammography. Special attention was focused on the milk ducts. All four quadrants of the breast and the retroareolar region were imaged systematically, and the presence and the width of the ducts were documented. After this examination, each woman was given a pain chart and a linear visual analogue scale (0 –100 mm, 0 ⫽ no pain, 100 ⫽ worst pain) to identify any breast pain and its intensity.17 Pain was recorded every day throughout the entire cycle. After judging the pain as cyclic or noncyclic according to Preece and colleagues,7 the day of the most intensive pain was assessed for the study. The patients were asked about recent typical symptoms of duct ectasia and periductal mastitis, such as nipple discharge, nonlactational infection, nipple retraction, and breast lump, and about smoking habits. Patients with breast pain were given another ultrasound scan during the period of the worst pain to correlate the site of the pain with the site of the dilated ducts. The sonographic picture was analyzed for a putative tissue edema. Homogeneous echogenicity of the parenchyma differing from fibrous structures in conjunction with the patient’s report of breast swelling was interpreted as tissue edema. Tissue edema could not be quantified. For these ultrasonographic scans, an AI 3200 (Dornier Instruments, Munich, Germany) was used with a linear array 7.5-MHz probe. This instrument can focus on different depths of the breast. Large breasts and small breasts could be examined with the same precision. The method successfully imaged the milk ducts in different parts of the breast. In some cases, a special maneuver or conformational change to the breast was necessary, as described by Harper and Kelly-Fry,18 to enable the ducts to be visualized. To clearly distinguish the milk ducts from the blood vessels, color Doppler sonography was applied. The lower threshold of this ultrasound technique in detecting a milk duct was a dilatation of more than 0.4 mm. In each patient, the milk ducts were identified, and their maximum width was evaluated. The
Peters et al
Mastalgia and Milk Duct Dilatation
55
Table 2. Patient Histories and Details of Study Findings Cyclical Noncyclical Asymptomatic mastalgia mastalgia (n ⫽ 123) (n ⫽ 136) (n ⫽ 76) Nipple discharge Nipple retraction Breast infection Lump Supposed tissue edema Nicotine addiction
2 (1.6)
5 (3.7)
2 (1.6)
9 (6.6) 63 (46.3)
13 (17.1) 1 (1.3) 2 (2.6) 3 (3.9) 3 (3.9)
14 (11.4)
20 (14.7)
37 (48.7)
Values are n (%).
ultrasound scans were exclusively done by one person (PD). The intraobserver variability of the width of the milk ducts was (mean ⫾ SD) 4.8% ⫾ 1.8% (n ⫽ 40). A prior study investigated whether there was any significant difference in the diameter of the milk ducts in the course of the menstrual cycle. Forty patients of the three groups with measurable milk ducts were randomly selected and given two separate ultrasound scans on days 7–10 and on days 24 –26 of the cycle. Statistical analysis was performed with the t test, the 2 test, correlation coefficient according to Pearson, and an analysis of variance. RESULTS In the entire study group, the width of the milk ducts measured ranged from 0.4 mm to 8.0 mm. Milk ducts
were undetectable in 21 women (⬍0.4 mm). One hundred twenty-three patients reported no symptoms in their breasts; 212 patients complained of pain. From a total of 212 mastalgia patients, 136 (64.2%) had cyclically pronounced and 76 (35.8%) had noncyclical mastalgia (Table 1). The mean age of the three groups (asymptomatic: 33.5 ⫾ 6.00 years, cyclically pronounced: 32.85 ⫾ 8.70 years, noncyclical: 34.32 ⫾ 6.69 years) did not vary statistically (analysis of variance). Details regarding clinical symptoms and nicotine addiction are listed in Table 2. Dilated milk ducts were found throughout all the quadrants of the breast, predominantly in the retroareolar region (Figure 1). In the 40 patients of the three groups with measurable milk ducts who were investigated during the first and the second part of the cycle, the width of the milk ducts did not vary significantly: follicular phase: 2.15 ⫾ 1.8 mm, luteal phase: 2.34 ⫾ 1.6 mm, P ⬎ .05 (t test for paired data). Therefore, the day of the cycle was not mentioned in this study. In asymptomatic women, milk duct widths from 0.6 mm up to 4.4 mm were measured. The mean maximum width was 1.8 ⫾ 0.84 mm in asymptomatic women (Table 1), 2.34 ⫾ 1.10 mm in cyclically pronounced (range 0.5–5.2 mm), and 3.89 ⫾ 1.26 mm (range 1.3– 8.0 mm) in noncyclical mastalgia (t210 ⫽ 9.30, P ⬍ .001). A tissue edema was diagnosed in 1.6% of asymptomatic women, in 46.3% of cyclical, and in 3.9% of noncyclical mastalgia patients (Table 2). Representative depictions
Figure 1. Allocation of milk ducts (⬎0.4 mm) to the particular quadrant and the areola of the breast in 64 patients. The circle in the top right corner of the figure describes the quadrants of the breast: 1 and 4 ⫽ outer quadrants, A ⫽ areolar region. Peters. Mastalgia and Milk Duct Dilatation. Obstet Gynecol 2003.
56
Peters et al
Mastalgia and Milk Duct Dilatation
OBSTETRICS & GYNECOLOGY
and 24 (17.7%) of the cyclically pronounced mastalgia patients pinpointed the pain at the site where the dilated ducts had been detected ultrasonographically (2 ⫽ 740.69, P ⬍ .001).
Figure 2. Representative ultrasonography of the breast of the three groups. a) Characteristic texture of a normal breast (asymptomatic) with clearly visible Cooper ligaments as well as breast and fat tissue between them. b) A narrow milk duct detectable behind the areola. c) Breast tissue of the upper outer quadrant of a patient with cyclical mastalgia. The normal structure (a) is covered with a “veil,” suggesting the clinically evident breast edema. d) Milk ducts in patients with cyclically pronounced mastalgia were found to be wider than in asymptomatic subjects. e) Diffused dilated milk ducts in a patient with noncyclical mastalgia (upper outer quadrant of the breast). f) A corresponding milk duct that is clearly dilated. Peters. Mastalgia and Milk Duct Dilatation. Obstet Gynecol 2003.
of the asymptomatic breast, premenstrual mastalgia, and noncyclical mastalgia with the corresponding milk ducts are presented in Figure 2. The severity of the pain (visual linear analogue scale) in noncyclical mastalgia (53.0 ⫾ 19.3 mm) significantly exceeded that of cyclically pronounced mastalgia (34.7 ⫾ 17.2 mm, t210 ⫽ 7.12, P ⬍ .001). The intensity of pain showed a significant, positive correlation with the width of the milk ducts (Figure 3, r ⫽ .5008, P ⬍ .001). Sixty-three (82.9%) of the noncyclical
VOL. 101, NO. 1, JANUARY 2003
DISCUSSION The data presented here indicate that dilatation of the milk ducts from a certain degree onward does indeed produce breast pain in nonlactating women. Regarding symptomatic mastalgia, duct ectasia is a major factor in this disorder, nearly exclusively so in noncyclical mastalgia, and it affects at least one third of the patients with cyclical mastalgia. Cyclical and noncyclical pain were evaluated together in the correlation analysis (Figure 1). A more correct analysis might be made by dividing the cases into breast pain groups; however, the aim of this study was to investigate a relationship between breast pain and the degree of dilatation of the ducts. The numerical distribution of women in the three groups neither reflects the actual incidence of mastalgia within the population nor the ratio between the two types of breast pain. The women were recruited from a cancer screening program, therefore patients with breast pain who were worried about having breast cancer are overrepresented. Furthermore, the asymptomatic group does not represent the normal population as control subjects. Therefore, bias is possible. There is a good correlation between the maximum width of the milk ducts and the degree of the pain documented. Patients with wider milk ducts have more pain than those with narrower ducts. The majority of our cases involved the terminal ducts and the subareolar region, sometimes only over a short distance. Moreover, the localized pain can be attributed to an ultrasonographically identified dilated duct at this site. In a noncyclical mastalgia group of patients, clinical examination by Griffith and colleagues found a palpable ectatic duct behind the nipple, with green- or yellow-stained nipple discharge or with a trigger zone pain (persistent and well localized).19 This correlates with our finding that ultrasonography reveals a dilated duct at the site of pain. Until now, the cause of mastalgia has not been clarified conclusively. In cyclical mastalgia, elevated prolactin response to thyrotropin-releasing hormone (TRH), and thus a higher daily prolactin production, has been documented.8 Bromocriptine as a potent prolactin-lowering drug is established in the treatment of cyclical mastalgia.12 A possible mechanism of action of prolactin may be fluid and electrolyte absorption. Although the effects of prolactin on osmoregulation in mammals are less clear, it has been shown that prolactin stimulates the retention of fluid and electrolytes in the intestine and
Peters et al
Mastalgia and Milk Duct Dilatation
57
Figure 3. Correlation between the width of the milk ducts and the intensity of breast pain (n ⫽ 168, r ⫽ 0.5008, P ⬍ .001). max ⫽ maximum. Peters. Mastalgia and Milk Duct Dilatation. Obstet Gynecol 2003.
kidney in rats.20,21 Moreover, prolactin may also play a role in osmoregulation in amniotic fluid by decreasing the diffusion of water from the fetal side to the maternal side of the membrane.22 Prolactin-related fluid retention within the breast may provide a possible explanation for premenstrual mastalgia. The increase of the breast volume during the premenstrual period is a well-known phenomenon.23 The tissue edema could be inferred from ultrasonography in some cases in our study, although it is difficult to quantify this phenomenon. Tissue edema may be responsible for a tension-mediated pain7,12,23 and may result in a compression of the milk ducts that are narrower than those in noncyclical mastalgia. Because breast secretion is thought to be controlled by prolactin,24 one would expect the wider milk ducts in patients with noncyclical mastalgia to be correlated with elevated prolactin levels. However, an earlier study by Kumar and colleagues,25 as well as a conjoined study by our group, could detect only normal basal and TRHstimulated prolactin values (manuscript in preparation). The response rate to prolactin-lowering treatment with bromocriptine is also less in patients with noncyclical mastalgia compared with patients with the cyclical type.12 Thus, cyclical and noncyclical mastalgia are probably two distinct entities. The reason why milk ducts are more distended in noncyclical mastalgia than in the cyclical type may at least be related to the tissue edema in the latter group. Because milk ducts also re-
58
Peters et al
Mastalgia and Milk Duct Dilatation
gress in width in postmenopausal women, attention should also be focused on estrogens. Further studies are needed to elucidate possible connections. Whatever the cause of distended milk ducts is, their manifestation is correlated with pain, as documented in our study. Milk ducts are known to be sensitive to pain. Every breastfeeding mother describes discomfort or pain of the filled breast before letting the child suckle. Breast engorgement during lactation can also be extremely painful. Therefore, the dilated milk ducts described by this study are most likely to be the origin of the pain, which is mediated by sensory nerve endings accompanying the blood vessels.26 Normal measurements for the diameter of a milk duct are not clearly defined. Regarding the histological criteria of duct ectasia with periductal mastitis, most authors describe milk ducts with a diameter of roughly 3 mm as fulfilling the definition.27 Based on our data, a cutoff value of 3 mm (eg, as a criterion for defining a dilated duct) is not clinically feasible in the same way. Most asymptomatic women have duct widths below this value, and most symptomatic patients have widths above it. Statistically, a threshold of maximum ductal dilatation that predicts breast pain cannot be evaluated on the basis of our data. The term duct ectasia is frequently associated with periductal mastitis. Two theories for its development have been formulated. The first proposes that milk ducts dilate with age and that stagnant secretion in these di-
OBSTETRICS & GYNECOLOGY
lated ducts leaks into the surrounding tissue to cause periductal mastitis.28 The second theory specifies that periductal inflammation predominates in younger women, whereas duct dilatation increases in frequency with advancing age.29,30 Periductal mastitis is frequently combined with clinical episodes of periductal mastitis and abscess formation.30 This especially affects women between the ages of 18 and 45.3,30 In our study, however, only two women reported a recent clinical mastitis, whereas typical symptoms of duct ectasia, such as nipple discharge and nicotine addiction,30 were mainly observed in patients with noncyclical mastalgia, in whom the greatest proportion of duct ectasia was seen. Breast biopsy was not indicated in the patients of this study. Consequently, we cannot report any data on periductal mastitis in noncyclical mastalgia. Because this clinical study found dilated ducts within an age group that is not typical for histologically proven duct ectasia,30 it seems possible that duct ectasia is a symptomatic variation of this condition that develops earlier in life. The reason milk ducts are dilated in patients with mastalgia is unknown; however, the study presented here provides a reasonable explanation for the general morphology underlying this disease. In conclusion, these results demonstrate for the first time that duct ectasia is a major factor in determining the severity of mastalgia. The site of pain is positively correlated with the site of duct dilatation in a considerable number of patients. REFERENCES 1. Venta LA. Image-guided biopsy of nonpalpable breast lesions. In: Harriss JR, Lippman ME, Morrow M, Osborne CK, eds. Diseases of the breast, 2nd ed. Philadelphia: Lippincott, Williams & Wilkins, 2000:149–64. 2. King TA, Carter KM, Bolton JS, Fuhrmann GM. A simple approach to nipple discharge. Arch Surg 2000;69:960–5. 3. Peters F, Kielich A, Pahnke V. Coincidence of nonpuerperal mastitis and breast cancer. Eur J Obstet Gynecol Reprod Biol 2002. In press. 4. Preece PE, Baum M, Mansel RE, Webster DJ, Fortt RW, Gravelle ICH, et al. Importance of mastalgia in operable breast cancer. BMJ 1982;248:1299–300. 5. Watt-Boolsen S, Emus H, Junge J. Fibrocystic disease and mastalgia. Dan Med Bull 1982;29:252–4. 6. Jorgensen J, Watt-Boolsen S. Cyclical mastalgia and breast pathology. Acta Chir Scand 1985;151:319–21. 7. Preece PE, Mansel RE, Bolten PM, Hughes LE, Baum M, Gravelle IH. Clinical syndromes of mastalgia. Lancet 1976;2:670–3. 8. Peters F, Schuth W, Scheurich B, Breckwoldt M. Serum prolactin levels in patients with fibrocystic breast disease. Obstet Gynecol 1984;64:381–5.
VOL. 101, NO. 1, JANUARY 2003
9. Goodwin PJ, Miller A, Del Giudice E, Singer W, Conelly P, Ritchie JWK. Elevated high-density lipoprotein cholesterol and dietary fat intake in women with cyclical mastopathy. Am J Obstet Gynecol 1998;179:430–7. 10. Peters F, Pickardt CR, Breckwold M. Thyroid hormones in benign breast disease: Normalization of exaggerated prolactin responsiveness to thyrotropin-releasing hormone. Cancer 1985;65:102–5. 11. Peters F. Multicentre study of gestrinone in cyclical breast pain. Lancet 1996;339:205–8. 12. Pye JK, Mansel RE, Hughes LE. Clinical experience of drug treatment for mastalgia. Lancet 1985;2:373–6. 13. Boyd NF, McGuire V, Shannon P, Cousins M, Kriukov V, Mahoney L, et al. Effect of a low-fat high-carbohydrate diet on symptoms of cyclical mastopathy. Lancet 1988;2: 128–32. 14. Wellings SR, Wolfe JN. Correlative studies of the histological and radiographic appearance of the breast parenchyma. Radiology 1978;129:299–306. 15. Moskowitz M, Gartside P, McLaughlin C. Mammographic patterns as markers for high risk benign breast disease and incident cancer. Radiology 1980;134:293–5. 16. Ayers J, Gidwani G. The “luteal breast” and hormonal and sonographic investigation of benign breast in patients with cyclical mastalgia. Fertil Steril 1983;408:779–84. 17. Tawaf-Mutamen H, Ader DN, Browne MW, Shriver CD. Clinical evaluation of mastalgia. Arch Surg 1998;133: 211–3. 18. Harper P, Kelly-Fry E. Ultrasound visualization of the breast in symptomatic patients. Radiology 1980;137: 465–9. 19. Griffith CDM, Dowle CS, Hinton CP, Blamey RW. The breast pain clinic: A rational approach to classification and treatment of breast pain. Postgrad Med J 1987;63:547–9. 20. Halloran BP, DeLuca HF. Calcium transport in small intestine during pregnancy and lactation. Am J Physiol 1992;239:64–8. 21. Stier CT, Cowden EA, Friesen HG. Prolactin and the rat kidney: A clearance and micropuncture study. Endocrinology 1984;115:362–7. 22. Tyson JE. The evolutionary role of prolactin in mammalian osmoregulation: Effects on fetoplacental hydromineral transport. Semin Perinatol 1982;6:216–28. ¨ ber Vera¨nderungen des Brustvolumens im 23. Do¨ring GK. U Cyclus. Arch Gynecol 1953;184:51–8. 24. Rose DP, Berke B, Cohen LA, Lahti H. A comparison of serum and breast duct fluid-immunoassayable prolaction and growth hormone with bioassayable lactogenic hormones in healthy women and patients with cystic breast disease. Cancer 1987;60:2761–5. 25. Kumar S, Mansel RE, Hughes LE, Woodhead JS, Edwards CA, Scanlon MF, et al. Prolactin response to thyrotropin-releasing hormone stimulation and dopami-
Peters et al
Mastalgia and Milk Duct Dilatation
59
26. 27.
28. 29.
60
nergic inhibition in benign breast disease. Cancer 1984; 153:1311–5. Vorherr H. The breast: Morphology, physiology and lactation. New York: Academic Press, 1974. Tedeschi L, Ahari S, Byrne J. Involutional mammary duct ectasia and periductal mastitis. Am J Surg 1963;106: 517–21. Haagensen CD. Mammary-duct ectasia: A disease which may simulate carcinoma. Cancer 1951;4:749–61. Bonser GM, Dossett JA, Jull JW. Human and experimental breast cancer. London: Pitman Medical, 1961:237–49.
Peters et al
Mastalgia and Milk Duct Dilatation
30. Dixon JM, Ravi Sekar O, Chetty U, Anderson TJ. Periductal mastitis and duct ectasia: Different conditions with different aetiologies. Br J Surg 1996;83:820. Address reprint requests to: Friedolf Peters, PhD, St. Hildegardis Hospital, Academic Teaching Hospital, Department of Obstetrics and Gynecology, D-55131 Mainz, Germany; E-mail: [email protected]. Received March 20, 2002. Received in revised form July 4, 2002. Accepted August 1, 2002.
OBSTETRICS & GYNECOLOGY