The presentation and outcomes of Hermansky- Pudlak syndrome in obstetrics and gynecological settings: A systematic review

Background: Hermansky- Pudlak syndrome (HPS) is a rare autosomal- recessive disor der with clinical manifestations of bleeding diathesis, multi- organ disease and variable oculocutaneous albinism (OCA). In women, it can cause life- threatening obstetric and gynecological (OB/GYN) bleeding. Objective: To summarize OB/GYN presentations, outcomes,


| INTRODUC TI ON
Hermansky-Pudlak syndrome (HPS) was first described in 1959. 1 It is characterized by dense granule deficient platelets (delta storage pool disorder), oculocutaneous albinism (OCA) with present tyrosinase activity and intracellular accumulation of ceroid lipofuscin.
Clinically, it manifests as variable degrees of bleeding diathesis, skin hypopigmentation, visual impairment, nystagmus, pulmonary fibrosis, granulomatous colitis and renal impairment. 2 It is heterogeneous in its presentation, clinical course and prognosis.
Early diagnosis in childhood is common due to OCA, which encompasses skin/eye hypopigmentation, visual impairment and nystagmus. Skin and hair color ranges from pale white to light brown/ tan and iris color tends to be blue, green or brown. 2,3 Cases of iris heterochromia are also reported. 4 Sometimes, OCA is only apparent after comparison to unaffected close relatives. HPS patients are prone to ultra-violet light mediated skin damage. 2,3 Bleeding symptoms of HPS, typical of platelet storage pool disorders, include excessive bruising, epistaxis, bleeding from other mucosal membranes, excessive bleeding after surgical or dental procedures, heavy menstrual bleeding (HMB) and postpartum hemorrhage (PPH). This is due to reduced platelet aggregation function resulting from the deficient dense granules/bodies. 2,3 Pulmonary fibrosis, inflammatory granulomatous colitis and renal failure are thought to be complications from multiple organ accumulation of an amorphous lipid-protein complex called ceroid lipofuscin. The lung pathology is a restrictive type with variable life expectancy, often presenting between the ages of 30-50 years. [5][6][7] Hermansky-Pudlak syndrome is a heterogeneous, autosomal recessive disorder and it is commonly found in Northwest Puerto Rico and the Swiss Valois region populations with an approximate prevalence of 1 in 1800 and a carrier frequency of 1 in 20. 3,8 Outside these populations it is rare, with a prevalence of about 1 in 100 000 to 1 000 000 worldwide. 9 Variants in at least 10 different genes are known to cause HPS (subtyped HPS1 to HSP10) of which variants in HPS1 are the most frequently reported. 9,10 The resultant intracellular defect is the disruption of the synthesis and trafficking of lysosome-related organelles, which include platelet dense bodies and melanosomes, thus impairing their relative functions. 2,10 The diagnosis of a delta storage pool disorder is indicated by low platelet nucleotides with decreased ADP:ATP ratio, reduced ATP secretion and characteristic impaired platelet aggregometry. 10 A definitive diagnostic feature seen under electron microscopy is the significant decrease in platelet dense granules/bodies. 11 Molecular genetic analysis enables classifications into the different subtypes of HPS and aids in management and prognosis. 5,10 Due to its rare nature, most of our knowledge on HPS is through case reports and mouse-model laboratory experiments. HPS presentation in obstetrics and gynecology (OB/GYN) settings can pose significant risks to the patient and challenges for the clinicians involved. This review collates and summarizes the presentations, different management strategies and their effects available in clinical case reports with the main focus on obstetrics and gynecological aspects of care.

| MATERIAL S AND ME THODS
A systematic literature search according to PRISMA guidelines 12 was performed to identify all eligible clinical cases and studies without language or time restriction. Electronic databases of MEDLINE, EMBASE, Cochrane, PubMed, Web of Science Core Collection and Google Scholar were searched from inception until June 30, 2020, using the following keywords: HPS, platelet storage pool disorders, OCA, women, females, pregnancy, labor, delivery, postpartum hemorrhage, obstetrics, HMB, menorrhagia, ovulation bleeding, ovarian bleeding and gynecology. The bibliographic references of the retrieved articles were also screened for additional cases/studies of relevance. We identified 220 articles altogether.
Abstracts of the articles in the literature searches were reviewed independently by two authors (DO-T and BH) for their relevance to OB/GYN outcomes in HPS patients. Sixteen articles were relevant and included in the review. The full articles were retrieved and the information extracted. One article was written in Spanish, so we obtained its translation into English. 13 The other 15 articles were published in English.
Quality assessment of case reports and case studies were evaluated using the adapted Newcastle-Ottawa Scale recommended by Murad  hemostatic prophylactic cover for menstruation; treatment for HMB and other gynecological problems and treatment) and obstetric (gravidity and parity; antenatal complications; hemostatic prophylactic cover for labor, delivery and puerperium; gestational age at delivery; mode of delivery (MOD); type of anesthesia employed; estimated blood loss (EBL) at delivery; postpartum complications and treatments received; any neonatal complications) data were also collected.

| RE SULTS
Of the 16 articles included in this review, 14 articles addressed pregnancies and two reported gynecological cases. There were 29 pregnancies described in 15 women and 2 gynecological cases in two women, making a total of 17 patients with HPS to evaluate. The data extracted from the cases are represented in Tables 1 and 2. 13,[15][16][17][18][19][20][21][22][23][24][25][26][27][28][29] In all the studies, diagnosis of HPS was confirmed by a hematologist; of which nine (56%) studies fully described the diagnostic criteria met by subject(s). Further subgroup analysis revealed a high percentage of studies (n = 15; 94%) adequately described and reported study outcome(s) and ensured adequate subject follow-up.
Whereas a lesser number of studies fully described patient baseline characteristics (n = 5; 31%) or adequately described the case(s) (n = 8; 50%). Consequently, reporting of patient baseline characteristics and adequate description of case(s) posed the greatest risk of publication bias.
The diagnosis of HPS in cases included in this review were made based on clinical and laboratory findings. Only two (12%) women had their diagnosis confirmed by genetic analysis but mutational information was not given. The ages at diagnosis were mentioned for 11/17 (65%) women and ranged from 6-42 years with a median age of 14 years.
In obstetrics, HPS diagnosis was known prior to first pregnancy in 9/15 (60%) women. Two (13%) women were diagnosed antenatally on clinical suspicion due to their OCA and bleeding diathesis history. PPH led to the diagnosis of HPS in three (20%) women. One (7%) woman presented after her fourth child with shortness of breath secondary to pulmonary disease, which led to her diagnosis. In gynecology, severe HMB necessitating packed red cells transfusion was the presenting symptom in both women that led to HPS diagnosis.
In two (12%) women, no information was given. The frequency of bleeding symptoms and other HPS associated signs/symptoms reported are described in Table 3. HMB was the most common bleeding symptom reported in 8/15 (53%) women. As part of OCA, nystagmus and significant visual impairment including blindness were reported in 11/15 (73%), 7/15 (47%) and 7/15 (47%) women, respectively.

| Gynecological presentation
The two gynecological cases were aged 13 and 42 years at presentation. They were both nulligravida and presented with HMB, which led to the diagnosis of HPS. Norethisterone was started at presentation with little effect. Her OCA roused a high suspicion for a platelet function disorder, so five doses of intravenous desmopressin (0.3 µg/kg every 8-12 h) and intravenous tranexamic acid (TXA, 12 mg/kg 3 times a day) were tried. However, desmopressin was not effective despite laboratory normalization of prolonged collagen/epinephrine PFA-100 closure times by it. Her HMB was resolved after 4 days with a single injection of recombinant factor VIIa (rFVIIa; NovoSeven ® at 100 µg/kg). Symptoms were managed with a progesterone derivative and TXA to prevent further HMB during periods. TXA was taken 2 days before her periods and continued 2 days after period cessation. She had a family history of parental consanguinity. Her younger sister was subsequently diagnosed with HPS due to OCA and skin bruising tendency. 28 The second case was in a 42-year-old woman of Indian ethnicity presenting with a 7-month history of HMB. She had skin hypopigmentation, easy bruising, nystagmus and poor visual acuity. Her hemoglobin level was 74 g/L and she was transfused packed red cells (number of units transfused not reported). She had a normal pelvic ultrasound. She was commenced on oral contraception, TXA and desmopressin, which controlled her heavy bleeding. She was continued on maintenance therapy of TXA and desmopressin during her periods. A chest X-ray and chest CT scan both showed mild pulmonary fibrosis. Her lung function was normal. An echocardiogram showed normal heart function. She had two siblings with similar phenotype and consanguineous parents. 29

| Pregnancy and obstetric outcomes
A total of 29 pregnancies were reported in 15 women; eight women had one pregnancy, three women had two pregnancies, two women had three pregnancies, one woman had four pregnancies and one woman had five pregnancies described. Twenty-seven pregnancies were singletons, one dichorionic diamniotic twin pregnancy and one dichorionic, triamniotic (DCTA) triplet pregnancy.
Two (7%) of the 29 pregnancies resulted in miscarriage. There was no further information regarding the management and bleeding outcomes of these miscarriages. Twenty-seven (27) viable pregnancies and deliveries were thus reported in 15 women.
The median age of the women during pregnancy was 22 years (age range 18-40 years). No antenatal bleeding was reported. None of the cases reported development of antiplatelet antibodies. One woman required bronchodilators in her pregnancy to relieve shortness of breath secondary to HPS-related pulmonary fibrosis. 22 The gestational age at delivery was mentioned in 23/27 (85%) cases. Twenty (87%) pregnancies delivered at term (i.e. ≥37 weeks gestation). The three preterm deliveries were in the same womanthe gestational ages at delivery for the first two pregnancies were not mentioned, but the third was the DCTA triplet pregnancy de-

| DISCUSS ION
This is the first extensive systematic review into the obstetric and gynecological problems and their management in women with HPS.
It highlights the bleeding diathesis during menstruation, labor and delivery associated with HPS, making affected women a high-risk population in obstetrics and gynecology setting.
The diagnostic age of HPS was wide, ranging from 6 to 42 years with a median age of 14, interestingly near the general age of menarche. 31 Thus, presentations may occur in a variety of medical specialties including pediatrics, OB/GYN, adult general medicine and surgery.
HMB was commonly the bleeding symptom reported, affecting 8 (53%) of women. Both gynecological cases initially presented with acute HMB and severe anemia. Therefore, platelet function disorders should be considered in the differential diagnosis of acute HMB. The treatment options include a combination of hormonal therapy and hemostatic therapy with TXA and/or desmopressin. Platelet transfusion and rFVIIa should be considered in cases of intractable bleeding.
Primary PPH occurred in 12 (44%) out of 27 deliveries, with half progressing into major PPH (EBL > 1000 ml). In three (20%) women, major PPH was the presenting symptom leading to diagnosis and they had the highest blood loss necessitating the use of high volumes of packed red cells, return to theatre and/or hemostatic products. 15,20,23 There were no reports of antenatal bleeding.
Desmopressin, platelet transfusion and/or TXA, were the main products used for both hemostatic prophylaxis and treatment in these patients, with platelet transfusion generally being preferred in high-risk hemorrhagic situations e.g. major surgery. 2,10 rFVIIa has been used to treat acute bleeding in patients with various platelet function defects with variable success. 28,32 In this review, we observed that platelet transfusion provided a better hemostatic cover in labor and delivery than desmopressin despite more cesarean sections in the platelet group. Comparing prophylaxis with desmopressin to platelet transfusion, the highest EBL was 1600 ml versus 800 ml and major PPH occurred in 3 (33%) versus none, respectively. In the prophylactic desmopressin Prolonged bleeding from minor cuts

(13)
Oral cavity bleeding 1 (7) Postoperative bleeding 1 (7) Petechiae 1 (7) a No information was given in two women. Non-bleeding complications of HPS requires consideration in managing these women. This review highlighted significant visual impairment and nystagmus in almost half the women (Table 3) Ideally, delivery should be in a center with obstetric experience and hemophilia unit with easy access to laboratory and blood products.
Planned delivery may be considered to facilitate the availability of resources and experienced clinicians.
The decision on type of hemostatic cover for delivery should take into consideration the woman's previous bleeding history, response to hemostatic treatment and her obstetric risk factors.
Desmopressin is safe in pregnancy and can be used initially; however, HLA-matched platelets should be ready if needed. Platelet transfusion should be used in women with a high risk of bleeding.
Active management of third stage of labor and steps to minimize tissue trauma during delivery in addition to appropriate hemostatic cover are important to minimize risk of PPH. Non-steriodal antiinflammtory drugs should be avoided to preserve any residual platelet function.
The MOD should depend mainly on obstetric indications. If the fetus is at risk of HPS, invasive procedures such as fetal blood sampling, fetal scalp electrode and instrumental deliveries should be avoided, except for easy lift out forceps when the fetal head is deeply engaged in the pelvis. The choice of analgesia/anesthesia should be individually assessed by the multi-disciplinary team.
Supportive measures like modulated lighting in care areas and increased healthcare workers' assistance for women with significant visual impairment should be established whilst hospitalized.