Should the modified natural cycle protocol for frozen embryo transfer be modified? A prospective case series proof of concept study.

OBJECTIVE: Modified natural cycles for frozen embryo transfer utilize an ovulation trigger which assists in embryo transfer scheduling and simplifies cycle monitoring. There have been conflicting results with this protocol and modifications may be sought. We wanted to ascertain whether a modified natural protocol for frozen embryo transfer without triggered ovulation but with luteal progesterone support disconnecting the timing of embryo transfer from the timing of the LH surge can achieve a high pregnancy rate. STUDY DESIGN: Candidates for frozen embryo transfer of 48-h cleavage cell embryos were recruited from May 2016 to April 2018. The patients were monitored for endometrial growth, follicle formation and estradiol, progesterone, and LH hormone levels. After meeting the predetermined criteria, embryo transfer was scheduled. The patients began progesterone treatment 48 h before embryo transfer, regardless of identification of the LH surge if ovulation had not commenced. The predetermined primary outcome was the biochemical pregnancy rate while the secondary outcome included the clinical pregnancy rate and the ongoing pregnancy rate. Patients were monitored to the eighth week of pregnancy, but data was collected from the medical records to provide the live birth rate as well. RESULTS: Fifty-six women were screened. Eleven women declined or did not meet the inclusion criteria. Three had anovulatory cycles and were excluded. Forty-two women were included in the statistical analysis. The implantation rate was 42.9 % [95 %CI 29.3 %-56.4 %). Of the 42 participants, 25 (59.5 %) conceived [95 % CI 44.0 %-75 %]. Two pregnancies ended in first trimester miscarriage leaving 23 (54.7 %) ongoing pregnancies [95 % CI 39.1 %-70.5 %]. One patient experienced a late abortion such that the live birth rate was 22 of 42 patients or 52.4 % [95 % CI 36.4 %-68.0 %]. CONCLUSION: The proposed modified natural protocol which utilizes progesterone luteal support but does not trigger ovulation, maintains a high pregnancy rate while providing flexibility regarding the day of transfer disconnected from the day of the LH surge. This was a prospective, proof of concept study. This protocol may be suitable for smaller or public in-vitro fertility units whose resources are limited and facilities are not available daily. The high pregnancy and live birth rate that we found provides confidence that this protocol can be part of the armament of protocols the clinician may offer to his patients. Larger studies should confirm these findings.

Lichenoid vulvar disease: A review.

Vulvar dermatoses are common, potentially debilitating conditions that can be seen by a variety of medical specialists. Lichenoid vulvar diseases, namely lichen sclerosus (LS), lichen planus (LP), and lichen simplex chronicus (LSC), can all negatively impact patients' quality of life and LS and LP also have an association with squamous cell carcinoma. It is essential that dermatologists are familiar with the unique features of each of these conditions to ensure the appropriate management and follow up. Herein, we provide an update on the epidemiology, clinical presentation, histopathology, and treatment of patients with vulvar LS, LP, and LSC.

Ectopic pregnancy risk factors for ART patients undergoing the GnRH antagonist protocol: a retrospective study.

BACKGROUND: In-vitro fertilization is a known risk factor for ectopic pregnancies. We sought to establish the risk factors for ectopic pregnancy in GnRH antagonist cycles examining patient and stimulation parameters with an emphasis on ovulation trigger. METHODS: We conducted a retrospective, cohort study of 343 patients undergoing 380 assisted reproductive technology (ART) cycles with the GnRH antagonist protocol and achieving a clinical pregnancy from November 2010 through December 2015. RESULTS: Significant risk factors for ectopic pregnancy in the univariate analysis included prior Cesarean section (CS), endometriosis, mechanical factor infertility, longer stimulation, elevated estradiol and progesterone levels, GnRH agonist trigger, higher number of oocytes aspirated, and insemination technique. Independent risk factors for ectopic pregnancy in the multivariate analysis included GnRH agonist trigger, higher number of oocytes aspirated, insemination technique, and prior Cesarean section. CONCLUSION: Excessive ovarian response, IVF (as opposed to ICSI), prior Cesarean section and GnRH agonist trigger were found to be independent risk factors for ectopic pregnancy. Caution should be exercised before incorporating the GnRH agonist trigger for indications other than preventing OHSS. When excessive ovarian response leads to utilization of GnRH agonist trigger, strategies for preventing ectopic pregnancy, such as a freeze all policy or blastocyst transfer, should be considered. Further studies should elucidate whether adjusting the luteal support can reduce the ectopic pregnancy risk.

A non-sense MCM9 mutation in a familial case of primary ovarian insufficiency.

Primary ovarian insufficiency (POI) results in an early loss of ovarian function, and remains idiopathic in about 80% of cases. Here, we have performed a complete genetic study of a consanguineous family with two POI cases. Linkage analysis and homozygosity mapping identified 12 homozygous regions with linkage, totalling 84 Mb. Whole-exome sequencing of the two patients and a non-affected sister allowed us to detect a homozygous causal variant in the MCM9 gene. The variant c.1483G>T [p.E495*], confirmed using Sanger sequencing, introduced a premature stop codon in coding exon 8 and is expected to lead to the loss of a functional protein. MCM9 belongs to a complex required for DNA repair by homologous recombination, and its impairment in mouse is known to induce meiotic recombination defects and oocyte degeneration. A previous study recently described two consanguineous families in which homozygous mutations of MCM9 were responsible for POI and short stature. Interestingly, the affected sisters in the family described here had a normal height. Altogether, our results provide the confirmation of the implication of MCM9 variants in POI and expand their phenotypic spectrum.

Empty follicle syndrome: successful treatment in a recurrent case and review of the literature.

Empty follicle syndrome is a condition in which no oocytes are retrieved after an apparently adequate ovarian response to stimulation and meticulous follicular aspiration. It is a rare condition of obscure etiology. A patient with primary infertility who underwent seven assisted reproductive technique cycles is described. In spite of a satisfactory ovarian response, aspiration yielded no oocytes in four cycles and 1-4 low quality oocytes in three cycles. In the index treatment cycle, ovulation was triggered using GnRH agonist 40 h prior to ovum pickup and hCG was added 6 h after the first trigger. Eighteen oocytes were recovered, of which 16 were mature and were inseminated by ICSI. Two embryos were transferred 48 h after aspiration and nine embryos were cryopreserved. The patient conceived and delivered a healthy boy at 38 weeks of gestation. The literature is reviewed and possible etiologies and treatment options of this enigmatic syndrome are suggested.

GnRH agonist therapy as ovarian protectants in female patients undergoing chemotherapy: a review of the clinical data.

BACKGROUND: Cancer survival rates in young women are improving due to progress in treatment. This includes aggressive chemotherapy, a treatment that often poses a threat to fertility. GnRH agonists were proposed as ovarian protectors during gonadotoxic therapies. This study was undertaken in order to determine the clinical evidence concerning this issue. METHODS: The medical literature was searched for studies that reported on ovarian function after the administration of GnRH agonists concomitant with chemotherapy. Twelve studies met the predetermined selection criteria. RESULTS: Data on ovarian function were obtained for 579 women who received chemotherapy. Among 345 women who received GnRH agonist co-treatment, ovarian function was preserved in 91% and 9% had premature ovarian failure. In 234 women who did not receive GnRH agonist co-treatment, ovarian function was preserved in 41% and failed in 59%. Only two of the studies were randomized. The control and the GnRH agonist groups differed in several important characteristics: the follow-up times were not equal, different treatment protocols were utilized and end-points were poorly defined and inconsistent between the studies. CONCLUSIONS: The effectiveness of GnRH agonists as fertility-preserving agents is debatable. A thorough literature search has found insufficient evidence to show that GnRH agonist co-treatment is effective in protecting the ovary from the damage of chemotherapy. A large randomized controlled trial with adequate follow-up is needed.

Prevalence and predictors of squamous cell abnormalities in Papanicolaou smears from women infected with HIV-1. Women's Interagency HIV Study Group.

BACKGROUND: Cervical neoplasia occurs with increased frequency among women infected with HIV-1. OBJECTIVE: To characterize prevalence of and risk factors for abnormal cervical cytology among women with HIV and to compare them to uninfected women. METHODS: Baseline cervical cytology was obtained from 1713 women seropositive for HIV and 482 at-risk control women who were enrolled in the Women's Interagency HIV Study, a multicenter prospective cohort study conducted in six U.S. cities. Associations with sociodemographic, medical, and sexual variables were assessed by Fisher's exact test, Mantel extension test, and logistic regression analysis. RESULTS: Cervical cytology was abnormal in 38.3% of HIV-infected women (atypical squamous cells of uncertain significance [ASCUS] 20.9%, low-grade squamous cells of uncertain significance [LSIL] 14.9%, high-grade squamous cells of uncertain significance [HSIL] 2.3%, cancer 0.2%) and 16.2% of HIV-uninfected women (ASCUS 12.7%, LSIL 2.3%, HSIL 1.2%, cancer 0.0%). Risk factors for any abnormal cytology in multivariate analysis included HIV infection, CD4 cell count, HIV RNA level, detection of human papillomavirus (HPV), a prior history of abnormal cytology, employment, and number of male sex partners within 6 months of enrollment. Prior abortion was associated with a decreased risk of cytologic abnormality. CONCLUSIONS: Cervical cytologic abnormalities were frequent among women infected with HIV, although high-grade changes were found in only 2.5%. Factors linked to sexual and reproductive history, HPV infection, and HIV disease all influenced risk.

Is HIV infection a risk factor for advanced cervical cancer?

OBJECTIVES: To compare HIV-infected and HIV-negative women with invasive cervical cancer with respect to predictors of advanced disease. METHODS: A retrospective analysis of 28 HIV-positive and 132 HIV-negative women with invasive cervical carcinoma was conducted and the two groups were compared with regard to stage of disease, demographic and behavioral variables, and risk factors for advanced disease. RESULTS: Overall, HIV-infected women were more likely to have advanced disease, because 78% of HIV-positive women had Stage II to IV compared with 55% of HIV-negative women (odds ratio [OR] = 3.1; p = .03). Substance abuse was strongly associated with HIV infection, as were high-risk sexual variables. Although HIV infection was associated with a threefold increase in advance stage cervical cancer in a univariate analysis, only symptom duration and lack of a recent Papanicolaou smear were significant predictors of advanced disease in a multiple logistic regression analysis. CONCLUSIONS: The major predictors of advanced cervical cancer are similar in HIV-positive and HIV-negative women, although the reasons for these predictors may be very different. It is likely that a large proportion of HIV-positive patients with cervical cancer acquire HIV infection after initiation of the neoplastic process.

Prevalence, risk factors, and accuracy of cytologic screening for cervical intraepithelial neoplasia in women with the human immunodeficiency virus.

OBJECTIVES: The objective was to evaluate the sensitivity and specificity of cervical cytology in women infected with the human immunodeficiency virus (HIV), risk factors for abnormal cytology in HIV-infected and uninfected women, and risk factors for histologic diagnosis of cervical intraepithelial neoplasia (CIN) in HIV-infected women. METHODS: Methods included a cross-sectional analysis of cervical cytology, colposcopic impression, and histology in 248 HIV-infected women and multivariate analyses of risk factors for abnormal cytology in 253 HIV-infected and 220 uninfected women and risk factors for CIN in 186 HIV-infected women. RESULTS: The sensitivity and specificity of cytology for all CIN grades were 0.60 and 0.80 and, for high-grade CIN, 0.83 and 0.74. The prevalence of abnormal cytology was 32.9% in HIV-infected and 7.6% in HIV-negative women. Independent risk factors for abnormal cytology were immunodeficiency [odds ratio (OR) 8-17, P < 0.001] and human papillomavirus (HPV) infection (OR = 5, P < 0.001). The prevalence of CIN on histology was 32% in HIV-infected women, and the only independent risk factor for CIN was oncogenic HPV type (OR = 5, P = 0.005). CONCLUSION: Given the high prevalence of abnormal cytology and CIN in HIV-infected women, cytologic screening has significant limitations. Both immunodeficiency and type of HPV infection are important risk factors.

Papillary serous and clear cell type lead to poor prognosis of endometrial carcinoma in black women.

OBJECTIVE: To investigate the significance of race and histologic type as prognostic factors in endometrial carcinoma. METHODS: We conducted a retrospective review of the medical records of all patients diagnosed with endometrial cancer from 1982 to 1995. Patients' clinical and pathologic characteristics were analyzed. RESULTS: The sample consisted of 401 patients, 59.9% (N = 229) were blacks and 40.1% (N = 153) were non-blacks. The mean age was 63.7 +/- 11.6 years. The histologic subtypes of endometrial carcinoma included 346 endometrioid (86.3%), 42 papillary serous (10.5%), and 13 clear cell (3.2%) adenocarcinomas. We found 79% of endometrioid adenocarcinomas were stage I or II compared to 26% of papillary serous tumors and 58% of clear cell carcinomas (P < 0.01). Eighty-eight percent of patients with papillary serous and 77% of patients with clear cell cancers were black (P < 0.01). Within each stage, patients were treated similarly irrespective of cell type or race. Five-year survival for endometrioid, papillary serous and clear cell adenocarcinomas was 69, 18, and 25%, respectively (P < 0.01). Black women had poorer 5-year survival (56%) than non-black women (71%). In multivariate analyses using age, stage, race, and histology, only stage and histology were independent risk factors for survival. CONCLUSIONS: Patients with papillary serous and clear cell endometrial cancer were more likely to be black, present at an advanced stage of disease, and have poor survival compared to patients with endometrioid adenocarcinoma. This may help to explain the poorer survival reported in blacks with endometrial cancer.

Uterine carcinosarcomas: incidence and trends in management and survival.

Our study's aim was to determine the incidence of uterine sarcomas in New York City (NYC) and evaluate trends in frequency, treatment, and survival of carcinosarcomas in two Brooklyn hospitals. Population-based cancer registry data for 1976-1985 were used to calculate the incidence of uterine sarcomas in NYC women. Medical records and histology slides of carcinosarcomas at two central Brooklyn hospitals from 1960 to 1995 were reviewed. The incidence of uterine sarcomas in black and white women in NYC was 33.4 and 17.0 per million (P < 0.01). Among 97 women with carcinosarcomas diagnosed in 1960-1995, 75% were diagnosed preoperatively, 82% had a hysterectomy, and 45% of those in clinical stage I were upstaged. Predictors of mortality included the presence of extrauterine extension, deep myometrial invasion, vascular space invasion, and gross residual disease, with only the first two being independent predictors of survival in a multivariate analysis. Adjunctive therapy shifted from radiation in 1960-1969 to cisplatin-based chemotherapy after 1980. In surgical stage III, survival increased significantly between 1960-1979 and 1980-1995, but improvement could not be ascribed to particular therapies. The incidence of uterine sarcomas in black women was twice that in white women. Surgical staging including omentectomy is recommended in the management of carcinosarcomas. Modern medical care may have improved the short-term prognosis of carcinosarcomas.

Vaginal intraepithelial neoplasia: risk factors for persistence, recurrence, and invasion and its management.

OBJECTIVE: Our purpose was to profile patients with vaginal intraepithelial neoplasia, evaluate the response to treatment and define risk factors for persistence and progression. STUDY DESIGN: We reviewed records and histopathology slides of 94 patients with vaginal intraepithelial neoplasia diagnosed from 1977 to 1986. For 74 patients with follow-up, we evaluated risk factors by univariate and multivariate analyses. RESULTS: Sixty-four of 94 patients (68%) had prior or concurrent anogenital squamous neoplasia, including 21 with invasive and 43 with intraepithelial. Twenty-three had prior radiotherapy, 10 had anogenital neoplastic syndrome, and 11 were immunosuppressed. In 52 of 74 treated patients (70%), vaginal intraepithelial neoplasia went into remission after a single treatment. In 18 patients (70%) vaginal intraepithelial neoplasia went into remission after a single treatment. In 18 patients (24%) recurrent vaginal intraepithelial neoplasia went into remission after chemosurgery, upper vaginectomy, or other treatments; in 4 (5%) it progressed to invasion. Significant multivariate risk factors for persistence or progression were multifocal lesions and anogenital neoplastic syndrome but not vaginal intraepithelial neoplasia grade, associated cervical neoplasia, or immunosuppression. CONCLUSIONS: Although most vaginal intraepithelial neoplasia goes into remission after treatment, 5% of cases may progress from occult foci to invasion in spite of close follow-up.

Cervical cancer as an AIDS-defining illness.

OBJECTIVE: To evaluate the importance of cervical cancer in the spectrum of human immunodeficiency virus (HIV)-related diseases at a single high-risk institution and to compare disease characteristics in HIV-infected women with cervical cancer and those with other AIDS-related malignancies. METHODS: We retrospectively reviewed data on cervical cancer and AIDS in women registered through the New York City Department of Health and institutional tumor registries from 1987 through 1995. RESULTS: During the study period, cervical cancer was diagnosed in 28 HIV-positive women. In 26, cervical cancer was the initial AIDS-defining illness, representing 4% (26 of 725) of the subjects, and it was the sixth most common initial AIDS-defining illness in women. Cervical cancer was the most common AIDS-related malignancy among women, representing 55% of the cases, followed by lymphoma (29%) and Kaposi sarcoma (16%). In 71% of the women with cervical cancer, HIV infection was diagnosed at the time of cancer presentation by routine testing, whereas in women with other malignancies, HIV diagnosis preceded cancer diagnosis (70%) by a mean of 2.7 years. Patients with other malignancies had greater immunosuppression (mean CD4 count 153/microL) than those with cervical cancer (mean CD4 count 312/microL). The recurrence rate for women with cervical cancer was 88%. Although the interval from cancer diagnosis to death was similar in all three groups (9.1-12.4 months), cancer was the cause of death in 95% of HIV-infected women with cervical cancer, compared with 60% of those with other AIDS-related malignancies. CONCLUSION: In urban populations at increased risk for both diseases, cervical cancer is an important AIDS-defining illness and may be the most common AIDS-related malignancy in women.

Age and race as prognostic factors in endometrial carcinoma.

This research was undertaken to evaluate the effects of age and race on prognosis of patients with endometrial carcinoma. A total of 279 patients with endometrial carcinoma treated at State University of New York-Health Science Center and Kings County Hospital Brooklyn, New York from 1975 to 1990 were retrospectively analyzed. Patients were arbitrarily divided into young and old groups (< or = 60 years or > 60 years old, respectively). The distribution of grade, clinical stage, and extent of myometrial invasion by age was determined for the entire group and for black and white patients, respectively. Young and old patients were stratified by clinical stage, grade, and extent of myometrial invasion. The corrected median survival of young and old patients by race was evaluated by Kaplan Meier's method of analysis. Older patients in general had higher clinical stage, higher grade, and greater depth of myometrial invasion than younger patients. Also, black patients had higher clinical stage, higher grade, and greater depth of myometrial invasion than white patients. Older black patients had the least favorable distribution of prognostic factors. Overall younger patients had a median survival of 200 months compared to 90 months for older patients (p = 0.0085). The overall corrected median survival for whites was 232 months compared to 108 months for blacks (p = 0.0001). The median survival of older black patients was worst at 40 months, compared to 155 months for older white patients (p = 0.0005). Age is a very important prognostic factor in endometrial carcinoma for both blacks and whites, and it appears to be more pronounced in older black patients.

Estrogen receptor immunocytochemistry in endometrial carcinoma: a prognostic marker for survival.

UNLABELLED: Only a few parameters such as tumor grade and stage are of value in prognosticating disease course in endometrial carcinoma. Biochemical steroid hormone receptor assays could also be useful but are difficult to perform and interpret. Immunocytochemical assay (ICA) might be the method of choice for detecting endometrial receptors. METHODS: Frozen tissue from 78 cases of endometrial adenocarcinoma was examined for the presence of estrogen (ER) and progesterone receptors (PgR) with specific monoclonal anti-receptor antibodies and the peroxidase-antiperoxidase method. In over 60 cases, frozen tissue was also assayed for ER and PgR by biochemical means. RESULTS: Fifty-five (71%) of the endometrial carcinomas were ERICA-positive and 55 (71%) PgRICA-positive. Although both ERICA and PgRICA correlated significantly with biochemical ER and PgR only ERICA was predictive of survival. A woman with a negative ERICA was 4 times more likely to die of her disease than if she were ERICA-positive (P = 0.009; mean follow-up, 37.5 months). Three cases ERICA-positive and PgRICA-negative survived while 3 others ERICA-negative and PgRICA-positive died. CONCLUSION: ERICA, a technique easy to perform and interpret at the community hospital level, appears to provide prognostic information independent of tumor stage and grade. Such information might be of value in planning postoperative therapies for women with endometrial cancer.

Multiple recurrences of cervical intraepithelial neoplasia in women with the human immunodeficiency virus.

OBJECTIVE: To evaluate the long-term outcomes after treatment of cervical intraepithelial neoplasia (CIN) in women infected with the human immunodeficiency virus (HIV). METHODS: Human immunodeficiency virus-infected and HIV-negative women treated for CIN by ablation or excision were followed-up prospectively by cytology and colposcopy for periods of up to 73 months. RESULTS: Among 127 HIV-infected CIN patients, 62% developed recurrent CIN by 36 months after treatment, compared with 18% of the 193 HIV-negative CIN patients. Recurrence rates reached 87% in 41 HIV-infected women with CD4 counts less than 200 cells/mm3. Progression to higher-grade neoplasia, including one invasive cancer, occurred by 36 months in 25% of HIV-infected and 2% of HIV-negative women. After adjusting for age, CIN severity, and treatment type, predictors of recurrence included HIV infection (rate ratio 4.4), and, in HIV-positive women, low CD4 count (rate ratio 2.2). In patients treated by excision, predictors of recurrence included HIV infection (rate ratio 2.0) and residual CIN after treatment (rate ratio 2.7). After a second treatment,a second CIN recurrence developed in 14 of 33 HIV-infected and in one of 17 HIV-negative women. After a third treatment, three of six HIV-infected women developed a third recurrence. With long-term follow-up, 45% of treated HIV-infected CIN patients had chronic condylomatous changes in the cervix compared with 5% of HIV-negative women. CONCLUSION: In HIV-infected women, CIN may recur despite multiple treatments, and chronic condylomatous changes are common. Innovative therapies for controlling CIN in HIV-infected women are needed.

Complications after treatment of cervical intraepithelial neoplasia in women infected with the human immunodeficiency virus.

OBJECTIVE: To compare the frequency of complications after treatment of cervical intraepithelial neoplasia (CIN) in human immunodeficiency virus (HIV)-infected and -seronegative women in an ambulatory setting. STUDY DESIGN: A retrospective record review of 15 HIV-infected and 44 HIV-negative women treated by laser therapy or cone biopsy and retrospective interviews of 20 HIV-infected and 44 HIV-negative women treated by cryotherapy. RESULTS: Four of 35 (11%) HIV-infected women had excessive bleeding after laser/cone or cryotherapy as compared to one of 88 (1%) HIV-negative women (odds ratio 11.27, P = .02). After laser/cone therapy, significantly more HIV-infected women (53%) had cervicovaginal infections than did HIV-negative women (18%). A higher prevalence of infection was associated with more severe immunodeficiency. CONCLUSION: HIV-infected women are vulnerable to complications after treatment of CIN and should be monitored closely.

CD4 lymphocytes in women with invasive and preinvasive cervical neoplasia.

OBJECTIVE: To assess the relationship between CD4 lymphocyte population and stage of disease in cervical neoplasia. METHODS: Study population was 107 women with invasive cervical cancer, 116 women with cervical intraepithelial neoplasia (CIN), and 32 women without neoplasia diagnosed in 1988-1994. All women under age 50 were seronegative for the human immunodeficiency virus (HIV). All women over age 50 with CD4:CD8 ratio below normal were HIV-negative. Stage was defined by FIGO criteria using clinical findings. CD4 and CD8 lymphocyte populations were enumerated by flow cytometry prior to treatment. The normal range of CD4 counts was defined as 537-1571 cells/mm3. RESULTS: Distribution of CD4 count was similar in stages I (n = 40), II (n = 24), and III (n = 32), with 31% below normal and 9% above normal (mean CD4 count = 881). However, in stage IV (n = 11), 64% were below normal and 18% above normal (mean CD4 = 591). The difference in distribution between stages I-III and stage IV was statistically significant. Among 116 CIN patients, 10% had CD4 counts below normal and 3% above normal (mean CD4 = 910). Among 32 women without cervical neoplasia, 0% had CD4 counts below normal and 3% above normal. The difference between CIN and invasive cancer in the distribution of CD4 counts and CD8 counts was significant (P < 0.01). There was no difference in the CD4 count distribution by CIN severity. Forty-five percent of patients with below-normal CD4 counts at diagnosis developed recurrent cancer compared to 43% of patients with normal or above-normal CD4 counts. CONCLUSION: Women with invasive cervical cancer have lower CD4 counts and a broader distribution compared to women with preinvasive or no neoplasia. Metastatic cancer at diagnosis was associated with severely depressed CD4 count.