Discussion
The case reported is interesting due to the unique association between NF1 and HIV infection, the complexity of genetic counseling, the possible effects of pregnancy on NF1, the possible complications of pregnancy, mode of delivery and complexity of anesthesia that needs a specialized team in a tertiary level hospital.
HIV infection has been reported only in few cases in non pregnant NF1 patients. de Castro et al (4) described a 38-year-old woman with NF1 and AIDS with peripheral T-cells lymphoma with good response to chemotherapy and Isa et al (5) a 30 years old woman with HIV infection whose disease remained stable with an adequate response to antiretroviral therapy. More recently Hiesgen and Variava (6) discussed about a 30-years-old woman with NF1 and somatostatinoma who died soon after diagnosis. Also Forte (7) described a case of a 41-year-old female patient with NF1 and uncontrolled HIV infection with simultaneously supratentorial and infratentorial pilocytic astrocytoma. At best knowledge of the Authors there is no case of NF1 pregnant woman with HIV infection described in the literature.
Prenatal genetic counseling may be complex because no phenotype-genotype correlation in NF1 patients and the rare association between NF1 and HIV. Many options are available. Preimplantation diagnosis is only possible trough planned pregnancy obtained by assisted reproduction but is performed only in some centers (Scoot et al 2019) (8). Villocentesis or amniocentesis may be proposed when viral load is undetectable or low, because there is no increase in the risk of fetal infection (Floridia et al 2017, Eppes 2017) (9,10). Furthermore the diagnosis of NF1 has been performed by evidence of NF1 paternal mutation detected in free DNA in maternal blood (Gruber et al 2018) (11). Request of prenatal diagnosis for NF1however seems correlated to many factor such as cultural background, educational level, previous child affected, longer follow-up, awareness on the natural history of NF1 with specific knowledges of the disease (Farhi et al 2008) (12) and many NF1 patients choose not to receive prenatal diagnosis (Terzi et al 2009 (13). Our couple refused prenatal diagnosis for NF1, because no willingness to interrupt pregnancy in an affected fetus by NF1.
Current data show that fertility in NF1 young women is not compromised, but miscarriage can occur frequently. Pregnancies have been reported in women with NF1, but concerns exist for the mother and the fetus. Stroke, acute vascular rupture, hypertension, preeclampsia, low platelet count, HELLP syndrome and eclampsia have been reported (Terry et al 2015, Leppavirta et al 2017) (14,15). Particularly hypertension due to pheochromocytoma or paragangliocytoma may change the prognosis for the mother and the fetus (Walther et al 1999) (16). Nevertheless, maternal mortality in compliant patient with NF1 is rare.
Fetal pathologies such as Intrauterine Growth Retard (IUGR), oligohydramnios, preterm delivery, sudden fetal death and emergency cesarean section have been described. Fetuses may also be affected by NF1 and may present higher incidence of malformations (Leppavirta et al 2018 (17). Our patient had two preterm deliveries, one due to abruptio placentae in the absence of hypertension and two out three child had clinical signs of NF1. Vaginal delivery wasn’t contraindicated due to normal pelvic examination and undetectable viral load. However in HIV patient during vaginal delivery fetal contact with maternal blood should be minimized and there are no definitive data if operative vaginal delivery may increase the risk of fetal infection (Peters et al 2017) (18).
The association of NF1 with HIV infection warrant an experienced multidisciplinary team. Generally general anesthesia should be avoided in pregnant women, but neuraxial anesthesia in asymptomatic spinal neurofibroma may be cause of specific concern in NF1 patients, because asymptomatic neurofibromas are not rare (Dounas the al 1995 et al 1999) (19). Before neuraxial anesthesia and analgesia, may be useful a MRI of the spine to detect such tumor that may contraindicate regional anesthesia (Spiegel 2005) (20). Furthermore it should be remembered that HIV infection with undetectable or low viral load is not a contraindication to local anesthesia (Gaisen 2003, Benton et Reese 2009, Gronwald et al 2011) (21,22,23).
Finally, an unknown issue is if natural pregnancy immunomodulation may increase the risk of malignancy when associated with HIV and NF1. It is known that 4-15% of NF1 patients, affected by plexiform neurofibromas, develop malignant degeneration during their life, with a latency period of about 10-20 years. Tumors usually arise before 39 years of age (Korf et al 2000) (24). This risk is cause of concerns. More rare tumors may affect NF1 patient such as pheochromocytoma, neuroblastoma and melanoma that arise from neural crest and also Wilms tumor, malignant nodular hidradenoma, leukemia, rhabdomyosarcoma, T-cell-lymphoma. Causes of high prevalence of neoplasm in NF1 patients, both benign and malignant, are not completely known, and genetic, environmental and hormonal factors may be implicated. Actually, it is unknown the possible effect, if any, of many pregnancies in the incidence of malignant degeneration in NF1 patients, but a relevant effect of pregnancy on tumor growth is well known, because up to 80% of skin tumors increases in dimensions in pregnancy (Schmutz 2003) (25). Moreover, during pregnancy many new tumors may be diagnosed, and up to 3% of NF1 patients are diagnosed during pregnancy due to the appearance of neurofibromas (Nebesio et al 2007, Cesaretti et al 2013) (26,27).
Our patient was HIV positive but maintained low viral loud and normal CD4 count throughout each pregnancy, showing no evident HIV immunosuppression so probably it did not affect significantly neurofibromas growth. However through seven year of observation no relevant transformation or malignant degeneration of neurofibromas was observed.