3D Whole-body skin imaging for automated melanoma detection.

BACKGROUND: Existing artificial intelligence for melanoma detection has relied on analysing images of lesions of clinical interest, which may lead to missed melanomas. Tools analysing the entire skin surface are lacking. OBJECTIVES: To determine if melanoma can be distinguished from other skin lesions using data from automated analysis of 3D-images. METHODS: Single-centre, retrospective, observational convenience sample of patients diagnosed with melanoma at a tertiary care cancer hospital. Eligible participants were those with a whole-body 3D-image captured within 90 days prior to the diagnostic skin biopsy. 3D-images were obtained as standard of care using VECTRA WB360 Whole Body 3-dimensional Imaging System (Canfield Scientific). Automated data from image processing (i.e. lesion size, colour, border) for all eligible participants were exported from VECTRA DermaGraphix research software for analysis. The main outcome was the area under the receiver operating characteristic curve (AUC). RESULTS: A total of 35 patients contributed 23,538 automatically identified skin lesions >2 mm in largest diameter (102-3021 lesions per participant). All were White patients and 23 (66%) were males. The median (range) age was 64 years (26-89). There were 49 lesions of melanoma and 22,489 lesions that were not melanoma. The AUC for the prediction model was 0.94 (95% CI: 0.92-0.96). Considering all lesions in a patient-level analysis, 14 (28%) melanoma lesions had the highest predicted score or were in the 99th percentile among all lesions for an individual patient. CONCLUSIONS: In this proof-of-concept pilot study, we demonstrated that automated analysis of whole-body 3D-images using simple image processing techniques can discriminate melanoma from other skin lesions with high accuracy. Further studies with larger, higher quality, and more representative 3D-imaging datasets would be needed to improve and validate these results.

Skin cancer risk among testicular germ-cell cancer survivors: a systematic review and meta-analysis.

Testicular germ-cell tumours (TGCT) are the most common cancer among young adult men. Previous studies suggested TGCT survivors have an increased risk for skin cancer. The goal of this study was to systematically review the literature and evidence regarding skin cancer risk among TGCT survivors compared with the general population. PubMed, EMBASE, Web of Science, Cochrane Databases and reference lists were included in the search. A systematic review of all comparative studies with more than 10 TGCT survivors reporting on skin cancer incidence was performed. A meta-analysis of the Standardized Incidence Rate (SIR) was calculated by pooling study-specific log-transformed estimates using the random-effects model. Risk of bias was assessed using the Newcastle-Ottawa Quality Assessment Scale. Nineteen studies that reported on 147 935 TGCT survivors were included. Pooled SIR for skin cancer and for melanoma incidence among TGCT survivors were 1.93 (95% CI 1.62-2.29, P < 0.0001) and 1.81 (95% CI 1.57-2.08, P < 0.0001), respectively. In conclusion, compared to the general population, TGCT survivors have an increased risk for developing skin cancer and melanoma. Additional long-term studies that include TGCT survivors, additional risk factors and all subtypes of skin cancer are required.

Vanishing congenital lung malformations: What is the incidence of true regression?

BACKGROUND: A congenital lung malformation (CLM) that is diagnosed on prenatal ultrasound exam may subsequently become undetectable on later scans, a "vanishing" CLM. OBJECTIVE: The purpose of our study is to characterize the prenatal natural history and postnatal outcomes of "vanishing" lesions treated at our institution. METHODS: We performed a retrospective chart review of 107 patients diagnosed prenatally with CLM at our institution. Comparisons were made using Kruskal-Wallis or t-test for continuous variables and Fisher's exact test or Chi-Square test for categorical variables. Multivariable analysis using logistic regression was performed. RESULTS: Of the 104 patients, 59 (56.7%) had lesions that became sonographically undetectable on serial ultrasound scans. Patients with lesions that vanished prenatally tended to need less Neonatal Intensive Care Unit (NICU) admission at birth (persistent CLM: 54.8%vs vanished CLM: 28.8%), decreased need for supplemental O2 at birth (persistent CLM: 31.0%vs vanished CLM: 11.9%), and decreased delay in feeds (persistent CLM: 26.2%vs vanished CLM: 8.5%) compared to those with persistent CLM. After multivariate analysis controlling for maternal steroid administration and sex, admission to NICU maintained a slight statistical significance, with patients in the vanishing CLM group 2.5 times less likely to be admitted to the NICU. None of our patients whose lesions vanished prenatally required mechanical ventilation. Eighty-six patients underwent postnatal computed tomography (CT) chest. Only 2 patients had lesions that regressed on postnatal CT. CONCLUSION: Lesions that vanish on prenatal imaging may be associated with improved clinical outcomes. The rate of true regression at our institution was as low as 2.3%.

Postexcision mammography is indicated after resection of ductal carcinoma-in-situ of the breast.

BACKGROUND: The adequacy of excision of ductal carcinoma-in-situ (DCIS) usually is confirmed with specimen mammography and histopathological assessment of specimen margins. Postexcision mammography of the involved breast is used at some centers. The objective of this study was to evaluate the impact of postexcision mammography in DCIS. METHODS: We conducted a retrospective chart review of all patients treated for DCIS at our institution from 1995 to 1998. RESULTS: Sixty-seven patients had postexcision mammography performed. Residual microcalcifications were identified in 16 patients (24%). Further surgery was precluded by precise mammographic-pathological correlation by using sliced-specimen mammography in two patients. Twelve patients had repeat wide excision, and two patients underwent mastectomy. Residual DCIS was identified at re-excision in 9 of 14 patients (64%). The margin status of the initial resection was negative in three of nine patients (33%) and positive or unknown in six of nine patients (67%). CONCLUSIONS: Postexcision mammography is a valuable technique that complements specimen mammography and histopathological margin assessment in confirming that an adequate excision of DCIS has been performed. Postexcision mammography should be performed in all patients with DCIS associated with mammographic calcifications who are treated with breast-conserving therapy.

A survey of pediatric malignancy in West Virginia.

The occurrence of malignancy in the pediatric age group is an uncommon but serious event. Since little data are available on the extent, nature or referral patterns of childhood cancer in West Virginia, we conducted a survey of 782 primary care physicians and 17 regional referral centers. The results showed that 249 cases of malignancy in the pediatric age group were reported and that 68% of children with newly diagnosed childhood malignancy were referred to institutions within West Virginia. We conclude that the incidence and distribution of types of malignancy in childhood in West Virginia parallels that of the nation, although there is some regional variation within the state.

The use of heparinase to neutralize residual heparin in blood samples drawn through pediatric indwelling central venous catheters.

Routine-coagulation screening is unreliable if obtained through a previously heparinized central venous catheter. Screening tests were performed on 14 paired peripheral and central venous catheter samples without and with heparinase. The heparinase treated central venous catheter samples correlated well with the peripheral samples and can be used to atraumatically screen for hemostatic abnormalities.

Curing children with leukemia in West Virginia.

Leukemia is the most common cancer in childhood with acute lymphoblastic leukemia (ALL) the most common subtype. While once uniformly fatal, today leukemia is a highly curable disease. To determine the outcomes of children with acute lymphoblastic leukemia in West Virginia, we performed a retrospective analysis of the results of treatment of children and adolescents with B-lineage ALL diagnosed between 2/86 and 1/91 and treated by the pediatric oncology teams at Morgantown or Charleston. Forty-one children with B-lineage ALL were identified and treated by a uniform protocol. Twenty-nine (71%) have remained disease-free for more than two years off therapy and are considered cured. Of the 10 patients who relapsed, five have now been off rescue therapy for greater than two years and are likely to be cured. Thirty-five of the original cohort of 41 children are alive and disease-free yielding an overall survival of 85%. The results of treatment of childhood leukemia in West Virginia are comparable to national data. Children with ALL diagnosed and treated by pediatric oncology teams in West Virginia have a very good chance of being cured.

Delirium. Quick recognition, careful evaluation, and appropriate treatment.

Delirium is a common medical condition, especially in elderly hospitalized patients. The syndrome is characterized by a short course of confusion and changes in perception and behavior. Early detection can be enhanced by routine assessment of cognitive functioning in hospitalized patients, especially those at risk for delirium. Prompt recognition and aggressive treatment of the underlying cause are essential for a positive outcome. Supportive measures are designed to calm and protect the patient and provide symptomatic relief until the precipitating condition is corrected.

Severe obstetric complications after aggressive treatment of Asherman syndrome.

Thirty-three cases of Asherman syndrome from Yale-New Haven Hospital are presented. Restoration of menses occurred in 91% of patients with pregnancy resulting in 79% of all treated patients. Spontaneous abortion occurred in only one patient (4%), whereas viable pregnancies resulted 96% of the time. Severe obstetric complications excluding premature labor occurred in three cases (12%) and included placenta increta, uterine sacculation, and a paper-thin uterine fundus (ie, uterine dehiscence). These cases are discussed to reinforce considering treated patients with Asherman syndrome to be a high-risk obstetric population, and to counsel and manage these patients accordingly.

Estrogen replacement therapy: current thinking and practice.

Radiographic evaluation of the skeleton is not needed to assess need for hormone replacement. No available method (including CT and dual photon absorptiometry) can predict osteoporosis. Because patients prone to osteoporosis cannot be preselected, all women should be considered probable victims. To avoid excess bone mineral loss, estrogens should begin soon after the diagnosis of menopause in women who do not have contraindications, regardless of symptoms.

Treatment of hot flashes with transdermal estradiol administration.

A randomized prospective double blind study was performed to assess the ability of a transdermal therapeutic system (TTS) delivering estradiol (E2) to suppress hot flashes (HFs) in symptomatic postmenopausal women. Patients were given placebo or E2 in four doses for a 20-day period, and serum gonadotropin and estrogen levels and the occurrences of HFs were measured. Administration of placebo had no measurable effect on either estrogen or gonadotropin levels or the occurrence of HFs. A dose-response relationship was found between the rate of E2 administered and the circulating level of E2, with 25, 50, 100, and 200 micrograms/24 h dosages raising the mean E2 concentrations from mean baseline levels of 5-8 pg/ml to 18, 38, 73, and 100 pg/ml, respectively. Estrone levels also increased with TTS application, but to a lesser extent than did E2 levels. Dose-response reductions of FSH and LH with increasing amounts of E2 administration occurred, but gonadotropin levels were not lowered in any of the patients into the ranges found in premenopausal women. TTS application significantly suppressed the occurrence of HFs at the 50 micrograms/24 h dosage and higher. A significant negative correlation (r = 0.6045; P less than 0.001) between E2 levels and the rates of occurrence of HFs was found during hormone administration. Based on this regression, 50% and 100% reductions of HFs should occur at 61 and 122 pg/ml E2. These data indicate that the transdermal delivery of E2 with these systems significantly reduced the occurrence of HFs and allowed definition of the therapeutic range of hormone replacement in terms of lost ovarian function, as reflected by circulating E2 levels.

Acute ovarian responses to a long-acting agonist of gonadotropin-releasing hormone in ovulatory women and women with polycystic ovarian disease.

A long-acting agonist of gonadotropin-releasing hormone (GnRH-a, 100 micrograms/day) was administered daily for 14 days in four patients with polycystic ovarian disease (PCOD) and eight ovulatory women (OW) to determine acute hormone responses. Initiation of GnRH-a treatment in OW on day 5 of their menstrual cycles (OW-day 5) stimulated a greater acute rise of serum follicle-stimulating hormone (FSH) than that seen in OW beginning treatment on day 2 (OW-day 2) or PCOD patients. FSH levels fell to baseline values with repeated injections, whereas luteinizing hormone levels remained elevated in all patients. An acute rise and progressive fall of estradiol (E2) was found in all groups. The OW-day 5 group demonstrated a secondary increase, which by day 14 was clearly greater than that found in the other groups. This secondary increase of E2 in the OW-day 5 group was associated with lower abdominal pain, whereas OW-day 2 and PCOD patients were asymptomatic. For comparison, human menopausal gonadotropin (150 IU/day for 3 days) stimulated a significantly greater increase of E2 in OW-day 5 than in PCOD patients. These studies indicate that daily GnRH-a administration induced variable effects on ovarian function, which depended on when it was begun during the menstrual cycle and whether it was given to ovulatory or PCOD subjects. In addition, abdominal discomfort associated with GnRH-a use in regularly OW can be avoided by commencing agonist administration earlier in their menstrual cycles.

LRFD6a has a dose-related stimulatory or inhibitory effect on the ovary in normal luteal phase women.

A dose response curve for LRFD6a-a potent LRF agonist-was established in normal midluteal phase women. At low doses (3-10 micrograms) an acute stimulatory effect on gonadotropins and ovarian steroidogenesis of estradiol and progesterone was observed. Despite large increases in circulating gonadotropins following the highest dose (30 micrograms LRFD6a), no acute changes in plasma steroids were noted. A sharp drop in progesterone was present after 3 days, with significant shortening of the cycle in 3 of 4 subjects in the group. While the stimulatory effect may be due to gonadotropin release and local action, the luteolytic effect observed is suggested to be a consequence of the direct inhibitory effect of the analog on ovarian function.

Pituitary hormones during the menopausal hot flash.

Eighteen postmenopausal women with severe hot flashes had continuous recordings of finger temperature and skin resistance as objective indexes of flushing episodes, and serial measurements of anterior pituitary hormones as indirect indexes of hypothalamic neurotransmitter activity. Significant increases of growth hormone, adrenocorticotropic hormone (ACTH), and luteinizing hormone (LH) occurred with maximal concentrations at 30, five, and 15 minutes, respectively, after the onset of the skin temperature rises. No significant fluctuations of prolactin (PRL), thyroid-stimulating hormone (TSH), or follicle-stimulating hormone (FSH) were observed. The mean serum cortisol concentration increased 15 minutes after the hot flash, presumably consequent to the preceding elevation of ACTH. Pituitary ACTH release may be secondary to hypothalamic cooling, whereas increased growth hormone and LH output and the thermoregulatory adjustments comprising the flushing episodes are all consistent with cyclic episodes of increased hypothalamic norepinephrine activity.

Direct action of androgen on hot flushes in the human male.

Hot flushes (HF) were demonstrated definitively in a hypogonadal man by continuous recordings of finger temperature, skin resistance and subjective symptoms. The magnitudes and temporal interrelationships of these physiological changes were similar to those previously observed in oophorectomized women. HF were abolished by injections of methyl testosterone (MT) and during oral administration of fluoxymesterone (FM), a non-aromatizable androgen. Increases of the non-SHBG bound fractions of T and E2 were noted during MT therapy, either of which may have been responsible for suppression of HF. No significant changes of the non-SHBG bound levels of E2 were observed during FM ingestion. These data confirm that typical HF occur in the male and support the concept that androgen acts directly on the central nervous system without requiring aromatization to estrogen in the brain or other tissues.

Origin of serum progestins in polycystic ovarian disease.

The glandular origin of excess circulating steroid hormones in women with polycystic ovarian disease has been difficult to establish with previously described perturbation techniques. Recently it was demonstrated that daily administration of a potent gonadotropin-releasing hormone agonist achieves complete and reversible suppression of ovarian steroid secretion. To examine the source of C-21 steroid hormones, circulating levels were measured before and after administration of the same agonist in polycystic ovarian disease subjects and normal control subjects. Serum levels of these hormones were also determined after administration of dexamethasone and adrenocorticotropic hormone (ACTH) as well as bilateral oophorectomy. Subjects with polycystic ovarian disease exhibited significant elevations of serum pregnenolone, 17OH -pregnenolone, and 17OH -progesterone by comparison with normal control subjects. The glandular origins of the excess levels of pregnenolone and 17OH -pregnenolone were more difficult to determine and appear to be different from that of 17OH -progesterone.

The incidence of late-onset congenital adrenal hyperplasia due to 21-hydroxylase deficiency among hirsute women.

Late-onset congenital adrenal hyperplasia (CAH) due to 21-hydroxylase deficiency is a cause of hirsutism in adult women. Its reported frequency of occurrence in hirsute women has varied from 0-30%, but the number of patients studied was small. To establish the incidence of CAH, 83 unselected hirsute women were studied prospectively with a standard ACTH stimulation test. On the basis of an exaggerated response of serum 17 alpha-hydroxyprogesterone to ACTH, 1 patient with CAH was found, for an incidence of 1.2%. The 95% confidence limits for the incidence of CAH among hirsute women were 0% and 3.4%. Five of seven hirsute women without CAH whose serum 17 alpha-hydroxyprogesterone levels rose above 3 ng/ml in response to ACTH had simultaneous serum progesterone values consistent with recent ovulation. Since routine screening of all hirsute women by means of ACTH stimulation does not appear to be cost effective, reported cases of CAH were reviewed in order to discern potentially helpful clinical clues. Severe hirsutism, virilization, early onset of symptoms, short stature, familial occurrence, and regular menses were identified as the clinical characteristics associated with late-onset CAH.

The effects of naloxone on hot flashes and gonadotropin secretion in postmenopausal women.

Hot flashes have a close temporal relationship with the initiation of LH pulses, suggesting that factors stimulating gonadotropin release are involved in the mechanism of this disturbance. It has been reported that the opiate antagonist naloxone acutely blocked subjective hot flashes, a seemingly paradoxical effect, since the use of this agent in premenopausal women increases the magnitude and frequency of LH pulses. We, therefore, studied the effects of naloxone in 16 postmenopausal women with frequent hot flashes using continuous recordings of finger temperature and skin resistance as objective indices of flushing and perspiration, respectively. After baseline recordings, the subjects were randomized into equal groups, and the recordings were repeated during 8-h infusion of either saline or naloxone (22 micrograms/min). Serum gonadotropin levels were measured at 15-min intervals before and during the last 4 h of the infusion. Naloxone did not change the rate of objectively measured hot flashes, mean serum LH or FSH levels, or the frequencies or amplitudes of gonadotropin pulses. These data suggest that there is a very low input of endogenous opiates on gonadotropin secretion in postmenopausal women and that opioid peptides do not play a role in the initiation of the postmenopausal hot flash.