Pregnancy-associated diabetes insipidus in Japan—a review based on quoting from the literatures reported during the period from 1982 to 2019

Abstract

This Review Article overviews the literature on diabetes insipidus (DI) associated with pregnancy and labor in Japan published from 1982 to 2019. The total number of patients collected was 361, however, only one-third of these cases had detailed pathophysiologic information enabling us to identify the respective etiology and subtype. Pregnancy-associated DI can be divided into 3 etiologies, central (neurogenic) DI, nephrogenic DI, and excess vasopressinase-associated DI. Neurogenic DI has various causes: for example, DI associated with tumoral lesions in the pituitary and neighboring area, DI associated with Sheehan’s syndrome and/or pituitary apoplexy, and DI associated with lymphocytic infundibuloneurohypophysitis (LINH, stalkitis). Nephrogenic DI results from defective response of the kidney to normal levels of vasopressin. However, the most interesting causal factor of pregnancy-associated DI is excess vasopressinase, caused either by excess production of vasopressinase by the placenta or defective clearance of vasopressinase by the liver. Hepatic complications resulting in pregnancy-associated DI include acute fatty liver of pregnancy (AFLP) and HELLP syndrome (syndrome of hemolysis, elevated liver enzymes, low platelets), as well as pre-existing or co-incidental hepatic diseases. A possible role of glucose uptake in putative stress-induced DI and the importance of correct diagnosis and treatment of pregnancy-associated DI, including use of 1-deamino 8-D arginine vasopressin, are also discussed.

Introduction

Diabetes insipidus (DI) is caused by deficient vasopressin (AVP, arginine vasopressin; also known as anti-diuretic hormone, ADH) responses resulting in excessive water loss and dehydration [1, 2]. DI associated with pregnancy is a rare endocrinopathy that can be caused by exacerbation during pregnancy of pre-existing DI or by decreased release of AVP by the posterior pituitary, defective response to AVP by the kidney, or by excessive vasopressinase-mediated degradation of AVP during pregnancy in patients without pre-existing DI [3-9].

In Japan, Dr. Taya was the first presenter to report a case of DI associated with pregnancy [10]. Since this report was made as an oral presentation, detailed records are not available, but we speculate that his case was a pre-existing DI exacerbated by pregnancy. In 1984, Kaneko et al. [11] analyzed 38 pregnancies associated with DI in Japan; 23 cases developed DI during pregnancy and recovered after parturition. Four cases of DI in pregnancy did not remit after delivery and one case became worse after delivery, possibly due to perinatal or postpartum complications that affected the posterior pituitary. Three years later, Higuchi et al. [12] reported a case of DI triggered by pregnancy and estimated that the number of cases of DI associated with pregnancy in Japan identified to that point in time was about 60. The present paper presents a summary of DI associated with pregnancy reported in Japan from 1982 to 2019, with brief descriptions of a few outstanding cases which could help our understanding of DI that develops during pregnancy.

Summary of case reports of pregnancy-associated diabetes insipidus in Japan from 1982 to 2019

Upon reviewing the published literature and the data therein available from the Japan Medical Abstract Society (JAMAS), the Japan Endocrine Society, the Japanese Society of Internal Medicine, and the Japan Society of Obstetrics and Gynecology, we found 361 cases (including our case report). On the basis of the total number of babies born during this period, the incidence of pregnancy-associated DI in successful pregnancies was calculated as 0.81 per 100,000 successful pregnancies (361/44,342,265) [13].

In the majority of the cases reported, pregnancy-associated DI followed the expected pattern of vasopressinase-mediated DI subtypes; DI develops later in pregnancy, as the levels of vasopressinase produced by the placenta are becoming elevated, and spontaneously remits postpartum, after delivery of the placenta. Both central DI and nephrogenic DI in pregnancy are generally responsive to the AVP analogue 1-deamino 8-D arginine vasopressin (DDAVP) [14, 15]. However, except for response to DDAVP, the majority of reports did not contain information concerning the pathological conditions of the patients. About one-third of the total reports did contain information pertaining to pituitary and liver pathologies in patients that developed pregnancy-associated DI (Table 1). These cases are discussed in further detail below.

Table 1 Pregnancy-associated DI with pituitary, liver and kidney pathologies

Pathophysiology	Number of cases
1. Pregnancy-associated DI due to pituitary injury:
a) Tumoral lesions of the pituitary and neighboring area	19
b) Sheehan’s syndrome	14
c) Pituitary apoplexy and bleeding	10
d) Lymphocytic hypophysitis, infundibuloneurohypophysitis and stalkitis	12
2. Pregnancy-associated DI due to liver pathologies:
a) Development of DI during pregnancy associated with acute fatty liver	15
b) Development of DI during pregnancy associated with HELLP syndrome	8
c) Development of DI during pregnancy associated with pre-existing liver or coincidental diseases	27
3. Nephrogenic diabeates insipidus complicated with pregnancy	12
Total	117

Brief overview of pathophysiological mechanisms of diabetes insipidus associated with pregnancy

Osmoregulation is markedly altered during normal human pregnancy. Decreased osmotic thresholds for thirst and AVP result in increased water intake and increased AVP expression and water retention [16], which among other effects results in increased blood volume and also softening of the body to allow for fetal growth. AVP is produced in the hypothalamus and is transported and released from the posterior pituitary. Central/neurogenic DI is due to impaired biosynthesis of AVP in the hypothalamus or impaired transport to and release from the posterior pituitary [1, 17, 18]; nephrogenic DI is due to lack of response of the kidney to AVP [1, 17, 18]; and vasopressinase-mediated DI is due to excess degradation of AVP by vasopressinase during pregnancy [3-7] (Fig. 1).

Fig. 1

Pregnancy-associated diabetes insipidus. The T-shaped dotted line denotes reduced hepatic inhibitory effect on vasopressinase activity. AVP, arginine vasopressin: LINH, lymphocytic infundibuloneurohypophysitis: HELLP, the hemolysis, elevated liver enzymes, low platelets syndrome

Pituitary. Retrieved February 23, 2021, from

https://ib.bioninja.com.au/options/option-d-human-physiology/d5-hormones-and-metabolism/pituitary-gland.html

Kidney. Retrieved February 23, 2021, from

https://www.123rf.com/photo_100657199_stock-vector-vector-set-of-healthy-realistic-human-organs-liver-spleen-heart-stomach-kidneys-and-lungs-isolated-o.html

Liver. Retrieved February 23, 2021, from

the University of Birmingham Web site:

https://www.birmingham.ac.uk/news/latest/2018/04/liver-transplant-patients-lives-saved.aspx

Placenta. Retrieved and partially modified February 23, 2021, from

https://www.kango-roo.com/ki/image_1134/

Presently vasopressinase is designated as Leucyl/cystinyl aminopeptidase (LNPEP) officially by Human Genome Organization Gene Nomenclature Committee. LNPEP has several nomenclatures such as oxytocinase, placental leucine aminopeptidase (P-LAP), insulin regulated aminopeptidase (IRAP), and angiotensin IV receptor (AT4) reflecting a broad range of substrates. Originally human vasopressinase had been identified and cloned as P-LAP from the placenta [19]. One year before, IRAP had been identified and cloned from rat fat tissue vp165 [20]. Subsequently it was discovered that P-LAP and IRAP were the homologues of the same protein. Furthermore, vasopressinase had been identified as AT4 by protein purification and sequencing of specific angiotensin IV binding site [21].

Oxytocin plays important roles in delivery, breastfeeding and maternal bonding behavior [22]. In particular, oxytocin is well known to induce strong uterine contractions. Vasopressinase is an insulin-regulated amino-peptidase expressed in syncytiotrophoblasts and increases during pregnancy as the placenta grows [16, 23, 24]. Vasopressinase degrades oxytocin as well as AVP, and the metabolic clearance rate of oxytocin increases markedly during pregnancy [25]. It has been suggested that vasopressinase may be involved in maintaining pregnancy homeostasis [26], and low levels of vasopressinase are associated with miscarriage and premature birth [27, 28]. In a normal pregnancy, increased synthesis and release of AVP is adequate to maintain functional AVP levels in the presence of the vasopressinase produced by the placenta [24]. However, in patients with pre-existing subclinical central or nephrogenic DI, the normal increase in vasopressinase that occurs during pregnancy can lower AVP levels sufficiently to exacerbate/unmask the disorder resulting in central or nephrogenic DI in pregnancy [29-32].

Patients with normal production and responses to AVP can develop DI during pregnancy if pregnancy-associated vasopressinase levels become excessive [16, 33, 34]. Patients with excess vasopressinase-associated DI can have up to 300 times higher levels of vasopressinase than normal healthy pregnant women [3, 4, 33, 35]. In these patients, degradation of AVP by vasopressinase decreases AVP to levels resulting in DI. Excess vasopressinase levels during pregnancy can be caused by excessive production of vasopressinase by the placenta [36] or by decreased degradation of vasopressinase in the liver [5, 6].

Liver diseases, including hepatitis (viral hepatitis, drug-induced hepatitis, and hepatitis of unknown etiology), acute fatty liver of pregnancy (AFLP), and HELLP syndrome (syndrome of hemolysis, elevated liver enzymes, low platelets), affect approximately 3% of pregnancies [37, 38]. Hepatic dysfunction can result in significant impairment of vasopressinase degradation [6, 32, 39, 40], and reports of DI in pregnancy due to excessive vasopressinase levels frequently cite an association of this form of DI with liver dysfunction and decreased metabolism of vasopressinase by the liver [35, 41-43].

Reports of excessive biosynthesis of vasopressinase are generally associated with multiple pregnancies [36, 44], however, other causes of increased vasopressinase production by the placenta have been suggested. Kondo et al. [45] reported a case of DI in pregnancy that they believed had been triggered by The 2016 Kumamoto Earthquake in Japan. They suggest that the earthquakes triggered stress-mediated release of oxytocin in the patient and premature uterine contractions. The placenta responded to the increased production of oxytocin by increasing synthesis of vasopressinase, causing the patient to experience DI.

Since the initial report of Blotner and Kunkel in 1942 [46] to later reports such as Krege et al. [41] and recent reports such as that by del Carpio-Orantes [47], it has been observed that DI in pregnancy generally occurs in the later stages of pregnancy and remits spontaneously a few days to a few months after delivery. However, in rare cases, it can occur after delivery, even after a Caesarean section [4, 6, 48]. Wallia et al. [48] reported a case in which placental abruption and subsequent release of placental vasopressinase resulted in postpartum DI. However, perinatal and postpartum development of DI is usually caused by perinatal or postpartum complications that affect the posterior pituitary, causing a subtype of central DI associated with parturition. A well-known cause of this subtype of DI associated with pregnancy is Sheehan’s syndrome, which occurs as a result of pituitary necrosis due to postpartum hemorrhage [49, 50].

Tumoral lesions of the pituitary and surrounding tissue affected by pregnancy, pregnancy-associated lymphocytic hypophysitis, and pregnancy-associated infundibuloneurohypophysitis can also cause DI.

Pregnancy-associated DI due to pituitary injury

We found 55 cases of pregnancy-associated DI that were caused by damage to the pituitary (Table 1). Nineteen cases were caused by damage to the pituitary and its stalk by cysts and tumors (Table 1). These cases included patients with invasion into the pituitary or oppression of the pituitary by prolactinomas (8 cases), pinealomas, non-functional pituitary adenomas, and other brain tumors and one patient with von Hippel-Lindau disease [51]. A well-known cause of another type of DI associated with pregnancy is Sheehan’s syndrome, which occurs as a result of pituitary necrosis due to postpartum hemorrhage, with DI developing within a few days after delivery [29, 50]. We found 14 cases of patients with Sheehan’s syndrome DI (Table 1). Patients with pituitary apoplexy and bleeding (10 cases), as well as twelve cases of lymphocytic hypophysitis and stalkitis were also collected (Table 1).

As noted above, DI associated with Sheehan’s syndrome occurs as a result of pituitary necrosis due to postpartum hemorrhage, and develops within a few days after delivery [29, 50]. However, an abstract entitled “Late onset postpartum diabetes insipidus: a case report” [52] appeared recently describing a case in which postpartum DI associated with Sheehan’s syndrome developed 6 months postpartum. The authors of the abstract note that it is not clear whether this case was simply coincidental or was triggered in the postpartum period by an unknown mechanism.

It is known that there are many cases having lymphocytic inflammatory lesions not only in the anterior pituitary (lymphocytic adenohypophysitis) but also in the posterior pituitary and pituitary stalk (lymphocytic infundibuloneurohypophysitis, stalkitis) [53, 54]. As in the general population, lymphocytic hypophysitis is the most common subtype of hypophysitis associated with pregnancy and parturition [55, 56]. Pregnancy/parturition-associated granulomatous hypophysitis has also been reported [57, 58]. Although the authors could find no reports of IgG4-related hypophysitis associated with pregnancy or the postpartum period in the literature, we noted 8 patients with pregnancy-associated hypophysitis and DI. Clinical features of these cases of pregnancy-associated DI related to lymphocytic hypophysitis are shown in Table 2 [59-66]. Among them, 4 cases developed lymphocytic hypophysitis in the postpartum period. Differing subtypes of lymphocytic hypophysitis is suggested by the fact that some cases of lymphocytic hypophysitis remit spontaneously while others lead to permanent hypopituitarism [53, 67]. Notably, lymphocytic hypophysitis is known to sometimes recur on pregnancy after a latent period as long as 5 years [63].

Table 2 DI during or after pregnancy complicated with lymphocytic hypophysitis and infundibuloneurohypophysitis

Case No	Age year-old	Initial symptoms	Onset of DI	Clinical and laboratory data	Image diagnosis	Therapy and effects	Ref.
1.	28 y	polydipsia, polyuria	36th week	elevated liver enzymes thrombocytopenia	MRI	C-section improved	[59]
2.	32 y	thirst, polyuria	32nd week	elevated liver enzymes prolonged prothrombin time elevated serum Na+	MRI	oxytocin-induced delivery DDAVP improved	[60]
3.	23 y	malaise, thirst, polyuria	35th week	elevated liver enzymes high bilirubin prolonged prothrombin time	MRI	transvaginal delivery DDAVP improved	[61]
4.	37 y	fatigue bitemporal anopia	24th week 28th week	low plasma cortisol low plasma thyroxine complication of IgA nephritis	MRI	C-section transsphenoidal surgery hormone replacement DDAVP	[62]
5.	34 y	headache, poor appetite	4 month postpartum	recurred infudibulo-hypophysitis abducens paralysis	MRI	transvaginal delivery steroid with DDAVP	[63]
6.	34 y	thirst, polyuria, visual disturbance	42nd week		MRI	C-section due to placenta abruption steroid with DDAVP improved	[64]
7.	32 y	headache, fatigue	32nd week	visual field defect biopsy revealed lymphocytic hypophysitis	CT	transvaginal delivery DDAVP with T4	[65]
8.	42 y	poor lactation, malaise, polydipsia, polyuria	after C-section in 34th week	hypothyroidism	MRI	C-section for hypertension under DDAVP and cortisol	[66]

week, gestational week; C-section, Caesarean section; DDAVP, 1-deamino 8-D arginine vasopressin

Conflicting results and poor specificity impaired the clinical usefulness of using anti-pituitary antibodies to diagnose hypophysitis. In 2015 Iwama et al. [68] identified rabphilin-3A as potentially useful marker of lymphocytic infundibuloneurohypophysitis and in 2017 Sakurai et al. [69] used antibodies against rabphilin-3A to diagnose central DI due to lymphocytic infundibuloneurohypophysitis that developed in the third trimester of a patient. However, the usefulness of rabphilin-3A antibodies for diagnosis of lymphocytic infundibuloneurohypophysitis remains to be confirmed. Currently, the most accepted diagnosis for hypophysitis and neighboring inflammatory lesions that can oppress the pituitary is careful reading of MRI images for homogenous swelling of these areas, including the pituitary stalk, compression of the floor of the sella turcica, and suprasellar extension of adenoma-like images [55]. Biopsy of the pituitary gland is the most reliable method for diagnosing the lymphocytic lesions, however, this method is invasive and is only undertaken in select cases in which diagnosis is uncertain and the results of the biopsy are likely to change the course of management of the disorder [55].

DI due to liver pathology associated with pregnancy

It is important to diagnose liver disease during pregnancy as early as possible and determine a relevant therapeutic course. Liver diseases in pregnancy are divided into 2 groups according to the recent article by Ma et al. [37]: (1) liver diseases specific to pregnancy and (2) liver diseases that are preexisting or coincident with pregnancy but not specific to pregnancy.

Group (1) in our case collection is the group of 15 patients with pregnancy-associated DI due to hepatic dysfunction caused by AFLP [70]. AFLP was described in 1940 as a specific clinical entity unique to pregnancy [71]. AFLP is a life-threatening complication of pregnancy with high maternal and fetal mortality. Symptoms include general malaise, fatigue, hypoglycemia, hypertension, nausea and/or vomiting, abdominal pain, jaundice, and skin pruritus [72, 73]. However, AFLP can be difficult to distinguish from HELLP syndrome [74]. The Swansea criteria [75] have a high negative predictive value for hepatic microvesicular steatosis, thus eliminating the need for liver biopsy in management of the disease [74, 76]. It is recommended that patients with AFLP have their pregnancies terminated upon diagnosis of the disease, either by Caesarean section or by forced labor [39, 74-76].

In our case collection (2), the onset-age of pregnancy-associated DI due to hepatic dysfunction caused by HELLP syndrome ranged from 26 to 39 years. The onset of initial symptoms ranged from the 32nd to 37th week of pregnancy. Minimal platelet count was 1.5 × 10⁴. A slight difference in the initial symptoms of the patients with HELLP syndrome from those with AFLP was a complaint of abdominal pain: epi-gastralgia. All cases underwent Caesarean sections, stillbirth was observed in two cases (Cases 1 and 5). It is noteworthy that the pathological factor of Case 3 was pituitary apoplexy. Another notable case was reported by Yamanaka et al. in 2002 [77]: a patient with twin pregnancies and suffering from HELLP syndrome developed DI during the postpartum period. The patient responded to administration of DDAVP. The authors hypothesize that the DI was due to somewhat increased vasopressinase production by the placenta due to the twin pregnancies coupled with decreased vasopressinase degradation by the liver due to hepatic dysfunction caused by HELLP syndrome [77].

Twenty-seven cases were also collected of pregnancy-associated DI due to preexisting liver diseases or liver diseases coincident with pregnancy but not specific to pregnancy (Table 1).

Case report

A 34-year-old patient was referred to the Department of Internal Medicine, KKR Maizuru Hospital, from the Department of Obstetrics and Gynecology (KKR Maizuru Hospital) because of marked polydipsia and pedal edema during her third trimester of pregnancy. Previously, she had an abortion at the 9th week of her first pregnancy. At that time, she had marked polydipsia and polyuria, as much as 10 liters a day. However, no abnormalities were noted during her annual health check-ups. One week after admission, the chief obstetrician decided to perform a Caesarean section, because of delayed labor. No serious complications occurred during the operation and a healthy male baby was delivered. Next morning, according to the opinion of the chief of the Internal Medicine, therapy using a nasal spray of DDAVP was begun, based on the data shown in Table 3. After delivery polydipsia and polyuria continued, suggesting that the patient’s pregnancy-associated DI was due to exacerbation of pre-existing central DI. Therefore, therapy with the DDAVP nasal spray was continued until DDAVP oral tablets became available. To investigate the possibility of hereditary neurogenic diabetes insipidus, mutational analysis was performed, however, no mutations in the patient’s AVP gene were found. Based on the absence of a high intensity signal in the posterior pituitary lobe on T1-weighted images, she was diagnosed to have central diabetes insipidus. The results of routine laboratory examinations, performed every 3 weeks after the operation, were within normal limits: serum electrolytes were low but near the normal limit, hepatic and renal function were normal. Also body weight remained constant at 63–64 kg under a nasal spray of DDAVP. Therefore, the patient was discharged 4 weeks after admission.

Table 3 Arginine-vasopressin test

	8 am		10:30 am		11 am		11:30 am		12:30 am
Urine volume (mL)		2,000		70		170		80
Serum Na (mEq/L)			153.7						152.1
Osmolaity (mOsm/kg)
Urine			68				382		403
Serum			296				294		295

Drinking water was prohibited from 8:00 am, and 5 units of water-soluble vasopressin (pitressin^®) were injected at 10:30 am.

Nephrogenic diabetes insipidus in pregnancy

Nephrogenic diabetes insipidus (NDI) is classified into either acquired or hereditary. Acquired NDI can occur secondarily to various renal diseases such as polycystic kidney disease, amyloidosis or sarcoidosis, and prolonged hypokalemia or hypercalcemia. Also those who prescribe or take medicine should be mindful that some medications such as lithium for emotional diseases, some antibiotics, and angiotensin II receptor antagonist can induce NDI.

Hereditary NDI is due to genetic defects causing malfunction of the kidney vasopressin receptors (AVPR2, X-linked, commonest) or mutations affecting the aquaporin-2 gene (AQP2, usually autosomal recessive).

Currently, options for the therapy of NDI are limited to the use of thiazide, NSAID (Non-Steroidal Anti-Inflammatory Drug), and potassium-sparing diuretics (Amiloride) [78].

Transient renal resistance to vasopressin during pregnancy is rare but well recognized. Pregnancy is a stress condition that unmasks insufficient abilities to secret and respond to vasopressin. Consequently, subclinical forms of NDI are exacerbated by increased vasopressinase activity and decreased reactivity to vasopressin during pregnancy [29, 79]. Since 1984, 12 cases of pregnancy-related NDI have been confirmed in Japan. Although not always clear, the causes or related abnormalities in each case include hypokalemia (4 cases) [31, 80-82], renal tubular injury (1 case) [80], drug (lithium, 1 case [83]; gentamicin, 1 case [15]), acute hepatic failure (1 case) [84], trisomy X (47, XXX) (1 case) [85], and NDI family history (2 cases) [29, 86].

The cause of hypokalemia is not always clear, but in one case, it was caused by refeeding syndrome following hyperemesis gravidarum [82]. A combination of hypokalemia and renal tubular injury was confirmed in one case [80]. In a patient with acute hepatic failure, the cause of NDI was unclear [84]. Three cases show recurrence of transient NDI during pregnancy [14, 29]. Another patient experienced both neurogenic and nephrogenic DI during pregnancy. The cause of NDI was unknown in this patient, but the coexistence of both neurogenic and nephrogenic DI suggested the possibility of a cystic lesion in the posterior pituitary [87]. A family history of NDI suggests mutations in either the AVPR2 or AQP2 genes. The clinical phenotype can be mild depending on the mutation [88, 89]. In the case of an AVPR2 mutation associated with the X chromosome, skewed X chromosome inactivation can make the clinical phenotype mild [90]. No genetic search has been performed, but patients with trisomy X or a family history are expected to have abnormalities similar to those described here.

A case of putative stress-induced DI associated with pregnancy

There is a report of a case of DI in pregnancy that was believed to have been triggered by The 2016 Kumamoto Earthquake. The authors [45] suggest that the earthquake triggered stress-mediated release of oxytocin in this patient, resulting in premature uterine contractions. The placenta might have responded to an upregulation of oxytocin by increasing synthesis of vasopressinase, thereby causing DI. In addition to the marked increase in vasopressinase in the placenta of the patient, Kondo et al. also found increased production of glucose transporter 4 (GLUT4) in the placenta, co-localization of GLUT4 with vasopressinase, and increased placenta weight [45]. They hypothesized that increased GLUT4 increased glucose uptake by the placenta, resulting in growth of the placenta, and that the relatively large placenta produced an increased amount of vasopressinase [45]. This is a fascinating case report and hypothesis, however, further investigation is indispensable.

Vasopressinase and Insulin

Vasopressin is also known as insulin regulated aminopeptidase (IRAP). The activity of IRAP promotes vesicles’ translocation to cell surface in response to insulin. Accordingly it is assumed that vasopressinase would improve insulin sensitivity. In accord with this premise, vasopressinase deficient mice showed diminished basal and insulin-stimulated glucose uptake in muscle and adipocytes [91].

Under basal conditions, IRAP localizes to the intracellular membrane compartments, and in response to insulin IRAP redistributes to the cell surface [92]. In type 2 diabetes, insulin production is inadequate and the response to insulin is also lower than normal, resulting in increased urine volume in response to increased blood sugar levels. However, decreased insulin responses also decrease the amount of vasopressinase translocated to the cell surface, and subsequently, the degradation rate of vasopressin would also be decreased. Thus, theoretically, increased vasopressin acts to depress the increased urine volume in type 2 diabetes patients.

Concluding remarks regarding our review of pregnancy-associated DI in Japan

Pregnancy-associated DI can be divided into 2 subtypes: one being vasopressinase-dependent and the other vasopressinase-independent. Vasopressinase is produced during pregnancy by the placenta, and its activity increases as the placenta grows during pregnancy and becomes undetectable at about 4–6 weeks postpartum [23]. Consequently, vasopressinase degradation of vasopressin and vasopressinase-mediated DI follow this pattern. Vasopressinase-independent DI occurs as a result of damage to the pituitary caused by perinatal and postpartum complications. Our collection of cases mostly followed the pattern expected of vasopressinase-mediated DI during pregnancy, with only about 15% (55/361) of the cases of pregnancy-associated DI resulting from pituitary damage. However, the overall incidence of pregnancy-associated DI cases we collected (approximately one in 100,000 successful pregnancies) is lower than expected (approximately 2 to 4 cases per 100,000 pregnancies). It is likely that this discrepancy resulted from under-assessment of pregnancy-associated DI. One factor that would have an effect on this underassessment is the fact that mild cases are likely to be under-diagnosed as thirst and polyuria are normal symptoms of pregnancy [1, 5, 6, 9, 35], and this under-diagnosis is not necessarily similar in different regions. Another factor is that we collected cases from abstracts and partial reports available from JAMAS, the Japan Endocrine Society, the Japanese Society of Internal Medicine, and the Japan Society of Obstetrics and Gynecology, and many of the abstracts and reports were too simplified, omitting essential data, particularly in the case reports presented at local meetings, to allow a confident diagnosis of pregnancy-associated DI. Consequently, many abstracts and reports were confined to the topics of the pathophysiological entities of pregnancy-associated DI, i.e., DI associated with lymphocytic hypophysitis, infundibuloneurohypophysitis (stalkitis), AFLP, and HELLP syndrome. Overall, our results are consistent with those from other regions.

Pregnancy-associated DI caused by excessive vasopressinase activity can be effectively treated with DDAVP. However, in our collection of cases, 3 stillborn births were documented, two stillborn births were from a mother suffering from AFLP and one stillborn birth was from a mother suffering from HELLP syndrome. These cases highlight the need to diagnose and treat liver dysfunction that occurs during pregnancy. In addition, work needs to continue to identify auto-antibodies that can help to diagnose cases of auto-immune hypophysitis.

Correct diagnosis and treatment, and prevention of pregnancy-associated DI are important in order to maintain healthy conditions during pregnancy and labor, for normal development of fetus, for ensuring healthy babies, and for contributing to earlier diagnosis and treatment of DI inherited by the next generation.

Acknowledgements

The authors are grateful to Professor David B. Alexander, Nagoya City University Graduate School of Science, for giving much kind and precise correction, also to Mr. K. Murata, Medical Library of Kanazawa University, and Ms. Y. Hida, a former clerk at Department of Internal Medicine, for helping collect the literature, to Dr. K. Kawahara with Mr. T. Katsumi for sending the patients’ records and to Mr. K. Hamaki and Mr. D. Nakamura, students at Kanazawa University School of Medicine, for helping organize the literature.

Disclosure

None of the authors has any potential conflicts of interest associated with this study.

References

• 1   Kalra  S,  Zargar  AH,  Jain  SM,  Sethi  B,  Chowdhury  S, et al. (2016) Diabetes insipidus: the other diabetes. Indian J Endocrinol Metab 20: 9–21.
• 2   Christ-Crain  M,  Bichet  DG,  Fenske  WK,  Goldman  MB,  Rittig  S, et al. (2019) Diabetes insipidus. Nat Rev Dis Primers 5: 54.
• 3   Ananthakrishnan  S (2009) Diabetes insipidus in pregnancy: etiology, evaluation, and management. Endocr Pract 15: 377–382.
• 4   Ananthakrishnan  S (2016) Diabetes insipidus during pregnancy. Best Pract Res Clin Endocrinol Metab 30: 305–315.
• 5   Chanson  P,  Salenave  S (2016) Diabetes insipidus and pregnancy. Ann Endocrinol (Paris) 77: 135–138.
• 6   Quigley  J,  Shelton  C,  Issa  B,  Sripada  S (2018) Diabetes insipidus in pregnancy. Obstet Gynaecol 20: 41–48.
• 7   Calina  D,  Docea  AO,  Golokhvast  KS,  Sifakis  S,  Tsatsakis  A, et al. (2019) Management of endocrinopathies in pregnancy: a review of current evidence. Int J Environ Res Public Health 16: 781.
• 8   Sert  M,  Tetiker  T,  Kirim  S,  Kocak  M (2003) Clinical report of 28 patients with Sheehan’s syndrome. Endocr J 50: 297–301.
• 9   Aleksandrov  N,  Audibert  F,  Bedard  MJ,  Mahone  M,  Goffinet  F, et al. (2010) Gestational diabetes insipidus: a review of an underdiagnosed condition. J Obstet Gynaecol Canada 32: 225–231.
• 10   Taya  M (1923) On a case of diabetes insipidus exacerbated with pregnancy. 9th Meet. Obstet. Gynecol. Soc. Kinki Dist. Japan 6: 1 (In Japanese).
• 11   Kaneko  N,  Oda  T,  Sekiguchi  Y,  Oono  T,  Komiya  Y, et al. (1984) Case studies on pregnancy and labor associated with idiopathic diabetes insipidus experienced and reported in this country. Clin Gynecol Obstet 38: 727–730 (In Japanese).
• 12   Higuchi  K,  Nakagawa  Y,  Miki  S (1987) Development of diabetes insipidus in a pregnant woman - a case report. IRYO—Japanese J Natl Med Serv 41: 372–378 (In Japanese).
• 13  Annual estimate of vital statistics for the first year of Reiwa (2019) | Ministry of Health, Labor and Welfare (Japan). https://www.mhlw.go.jp/toukei/saikin/hw/jinkou/suikei19/index.html accessed on October 28 (In Japanese).
• 14   Takemura  M (1995) A case in which DDAVP was effective for this gestational diabetes insipidus after repeating gestational nephrogenic diabetes insipidus twice in the past. Adv Obstet Gynecol 47: 494 (In Japanese).
• 15   Kotani  Y,  Mtsui  H,  Kawahara  K,  Muraoka  M,  Takagi  K (2012) A case of transient nephrogenic diabetes insipidus that developed after cesarean delivery, which is thought to be triggered by treatment for skin lesions at the extremities. Tokyo J Obstet Gynecol 61: 87–92 (In Japanese).
• 16   Lindheimer  MD,  Davison  JM (1995) Osmoregulation, the secretion of arginine vasopressin and its metabolism during pregnancy. Eur J Endocrinol 132: 133–143.
• 17   Di Iorgi  N,  Napoli  F,  Allegri  AEM,  Olivieri  I,  Bertelli  E, et al. (2012) Diabetes insipidus—diagnosis and management. Horm Res Paediatr 77: 69–84.
• 18   Kochhar  RS,  Ball  S (2017) Diabetes insipidus. Med (United Kingdom) 45: 488–491.
• 19   Rogi  T,  Tsujimoto  M,  Nakazato  H,  Mizutani  S,  Tomoda  Y (1996) Human placental leucine aminopeptidase/oxytocinase: a new member of type II membrane-spanning zinc metallopeptidase family. J Biol Chem 271: 56–61.
• 20   Keller  SR,  Scott  HM,  Mastick  CC,  Aebersold  R,  Lienhard  GE (1995) Cloning and characterization of a novel insulin-regulated membrane aminopeptidase from Glut4 vesicles. J Biol Chem 270: 23612–23618.
• 21   Albiston  AL,  McDowall  SG,  Matsacos  D,  Sim  P,  Clune  E, et al. (2001) Evidence that the angiotensin IV (AT4) receptor is the enzyme insulin-regulated aminopeptidase. J Biol Chem 276: 48623–48626.
• 22   Yamamoto  Y,  Liang  M,  Munesue  S,  Deguchi  K,  Harashima  A, et al. (2019) Vascular RAGE transports oxytocin into the brain to elicit its maternal bonding behaviour in mice. Commun Biol 2: 76.
• 23   Davison  JM,  Sheills  EA,  Barron  WM,  Robinson  AG,  Lindheimer  MD (1989) Changes in the metabolic clearance of vasopressin and in plasma vasopressinase throughout human pregnancy. J Clin Invest 83: 1313–1318.
• 24   Schrier  RW (2010) Systemic arterial vasodilation, vasopressin, and vasopressinase in pregnancy. J Am Soc Nephrol 21: 570–572.
• 25   Thornton  S,  Davison  JM,  Baylis  PH (1990) Effect of human pregnancy on metabolic clearance rate of oxytocin. Am J Physiol—Regul Integr Comp Physiol 259: R21–R24.
• 26   Nomura  S,  Ito  T,  Yamamoto  E,  Sumigama  S,  Iwase  A, et al. (2005) Gene regulation and physiological function of placental leucine aminopeptidase/oxytocinase during pregnancy. Biochim Biophys Acta—Proteins Proteomics 1751: 19–25.
• 27   Kozaki  H,  Itakura  A,  Okamura  M,  Ohno  Y,  Wakai  K, et al. (2001) Maternal serum placental leucine aminopeptidase (P-LAP)/oxytocinase and preterm delivery. Int J Gynecol Obstet 73: 207–213.
• 28   AL Sadoon  TH,  Al-Salihi  FG,  Al-Badran  AI (2014) Placental leucine aminopeptidase/oxytocinase expression in miscarriage. Br J Med Med Res 4: 3283–3292.
• 29   Iwasaki  Y,  Oiso  Y,  Kondo  K,  Takagi  S,  Takatsuki  K, et al. (1991) Aggravation of subclinical diabetes insipidus during pregnancy. N Engl J Med 324: 522–526.
• 30   Williams  DJ,  Metcalfe  KA,  Skingle  L,  Stock  AI,  Beedham  T, et al. (1993) Pathophysiology of transient cranial diabetes insipidus during pregnancy. Clin Endocrinol (Oxf) 38: 595–600.
• 31   Jin-No  Y,  Kamiya  Y,  Okada  M,  Watanabe  O,  Ogasawara  M, et al. (1998) Pregnant woman with transient diabetes insipidus resistant to 1- desamino-8-D-arginine vasopressin. Endocr J 45: 693–696.
• 32   Soule  SG,  Monson  JP,  Jacobs  HS (1995) Case report: transient diabetes insipidus in pregnancy—a consequence of enhanced placental clearance of arginine vasopressin. Hum Reprod 10: 3322–3324.
• 33   Durr  JA,  Hoggard  JG,  Hunt  JM,  Schrier  RW (1987) Diabetes insipidus in pregnancy associated with abnormally high circulating vasopressinase activity. N Engl J Med 316: 1070–1074.
• 34   Eglin JM  Jr,  Jessiman  AG (1959) Inactivation of the antidiuretic activity of vasopressin during pregnancy: a report of a case. J Clin Endocrinol Metab 19: 369–374.
• 35   Durr  JA,  Lindheimer  MD (1996) Diagnosis and management of diabetes insipidus during pregnancy. Endocr Pract 2: 353–361.
• 36   Cardino  MJT,  Isip-Tan  I (2012) Gestational diabetes insipidus in a 21 year-old G2P1 with twin pregnancy. Philipp J Intern Med 48: 33–37.
• 37   Ma  K,  Berger  D,  Reau  N (2019) Liver diseases during pregnancy. Clin Liver Dis 23: 345–361.
• 38   Ryan  JM,  Heneghan  MA (2014) Pregnancy and the liver. Clin Liver Dis 4: 51–54.
• 39   Ananthakrishnan  S (2020) Gestational diabetes insipidus: diagnosis and management. Best Pract Res Clin Endocrinol Metab 34: 101384.
• 40   Marques  P,  Gunawardana  K,  Grossman  A (2015) Transient diabetes insipidus in pregnancy. Endocrinol Diabetes Metab Case Rep 2015: 150078.
• 41   Krege  J,  Katz  VL,  Bowes  WA Jr (1989) Transient diabetes insipidus of pregnancy. Obstet Gynecol Surv 44: 789–795.
• 42   Kennedy  S,  Hall  PM,  Seymour  AE,  Hague  WM (1994) Transient diabetes insipidus and acute fatty liver of pregnancy. BJOG An Int J Obstet Gynaecol 101: 387–391.
• 43   Sainz Bueno  JA,  Villarejo Ortíz  P,  Hidalgo Amat  J,  Caballero Fernández  V,  Caballero Manzano  M, et al. (2005) Transient diabetes insipidus during pregnancy: a clinical case and a review of the syndrome. Eur J Obstet Gynecol Reprod Biol 118: 251–254.
• 44   Abbasi Ranjbar  Z,  Bahrami  A,  Najafipoor  F,  Mobassery  M (2010) Transient diabetes insipidus in a woman with a twin pregnancy: a case report. Int J Endocrinol Metab 8: 94–96.
• 45   Kondo  T,  Nakamura  M,  Kitano  S,  Kawashima  J,  Matsumura  T, et al. (2018) The clinical course and pathophysiological investigation of adolescent gestational diabetes insipidus: a case report. BMC Endocr Disord 18: 4.
• 46   Blotner  H,  Kunkel  P (1942) Diabetes insipidus and pregnancy – reports of two cases. N Engl J Med 227: 287–292.
• 47   Del Carpio-Orantes  L (2017) Transient central diabetes insipidus associated with pregnancy. Case report and bibliography review. Rev Med Inst Mex Seguro Soc 55: 672–675 (In Spanish).
• 48   Wallia  A,  Bizhanova  A,  Huang  W,  Goldsmith  SL,  Gossett  DR, et al. (2013) Acute diabetes insipidus mediated by vasopressinase after placental abruption. J Clin Endocrinol Metab 98: 881–886.
• 49   Iwasaki  Y,  Oiso  Y,  Yamauchi  K,  Takatsuki  K,  Kondo  K, et al. (1989) Neurohypophyseal function in postpartum hypopituitarism: impaired plasma vasopressin response to osmotic stimuli. J Clin Endocrinol Metab 68: 560–565.
• 50   Keleştimur  F (2003) Sheehan’s syndrome. Pituitary 6: 181–188.
• 51   Tsuji  K,  Fujiwara  H (2010) A case of von Hippel-Lindau disease presenting with transient central diabetes insipidus at the end of pregnancy. Folia Endocrinol Jpn 86: 108 (In Japanese).
• 52   Tuna  MM,  Doğan  BA,  Küçükler  K,  İmga  NN,  Karakılıç  E, et al. (2014) Late onset postpartum diabetes insipidus: a case report. Endocr Abs 35: P872
• 53   Imura  H,  Nakao  K,  Shimatsu  A,  Ogawa  Y,  Sando  T, et al. (1993) Lymphocytic infundibuloneurohypophysitis as a cause of central diabetes insipidus. N Engl J Med 329: 683–689.
• 54   Leggett  DAC,  Hill  PT,  Anderson  RJ (1999) “Stalkitis” in a pregnant 32-year-old woman: a rare cause of diabetes insipidus. Australas Radiol 43: 104–107.
• 55   Gubbi  S,  Hannah-Shmouni  F,  Verbalis  JG,  Koch  CA (2019) Hypophysitis: an update on the novel forms, diagnosis and management of disorders of pituitary inflammation. Best Pract Res Clin Endocrinol Metab 33: 101371.
• 56   Gutenberg  A,  Hans  V,  Puchner  MJA,  Kreutzer  J,  Brück  W, et al. (2006) Primary hypophysitis: clinical-pathological correlations. Eur J Endocrinol 155: 101–107.
• 57   Joneja  U,  Hooper  DC,  Evans  JJ,  Curtis  MT (2016) Postpartum granulomatous hypophysitis: a case study, review of the literature, and discussion of pathogenesis. Case Rep Pathol 2016: 7510323.
• 58   Youssef  A,  Ali  R,  Marouf  M (2018) Granulomatous hypophysitis in a postpartum patient: a case report. Cogent Med 5: 1428034.
• 59   Okazaki  M,  Kaji  H,  Kuboki  T,  Morimoto  I,  Baba  Y, et al. (1999) A suspected case of lymphocytic infundibuloneurohypophysitis, complaining of marked polyuria in the late stage of pregnancy. Pharma Medica 17: 238 (In Japanese).
• 60   Tamaru  N,  Yoshizawa  K,  Watanabe  F,  Fukuyama  J (2002) A case of DI, developed in the 36th week of pregnancy, showed lymphocytic infundibuloneurohypophysitis-like image on MRI, and spontaneously remitted. Horumonn-to-Rinsho (Clinical Endocrinol) 50: 26–29 (In Japanese).
• 61   Yamane  K,  Sato  S,  Nakayama  H,  Kira  N,  Yamaguchi  Y, et al. (2009) A case of lymphocytic hypophysitis developed in the late stage of pregnancy. Progr 61st Annu Congr Japan Soc Obstet Gynecol 61: P1-196 (Abstract) (In Japanese).
• 62   Arai  Y,  Nabe  K,  Ikeda  H,  Honjo  S,  Wada  Y, et al. (2007) A case of lymphocytic panhypophysitis (LPH) during pregnancy. Endocrine 32: 117–121 (In Japanese).
• 63   Miyata  M,  Midorikawa  S,  Honma  M,  Muraoka  R,  Kusano  H, et al. (2007) A case of lymphcytic hypothalamohypophysitis developed after 5 years, triggered for pregnancy. J Japan Endocrin Soc 83: 206–207 (In Japanese).
• 64   Mori  M,  Morikawa  M,  Sakai  K,  Mitsui  T,  Nakamura  H, et al. (2007) A case of lymphocytic hypophysitis developed diabetes insipidus and hypopituitarism due to massive postpartum bleeding. Progr 59th Annu Congr Japan Soc Obstet Gynecol 59: P2-483 (Abstract) (In Japanese).
• 65   Ota  K,  Kimura  T,  Hanyu  K,  Murakami  O,  Shoji  M, et al. (1993) A case of lymphocytic hypophysitis associated with hypopituitarism and cranial diabetes insipidus in the late stage of pregnancy. Horumon-to-Rinsho (Clinical Endocrinol) 4: 61–67 (In Japanese).
• 66  Fukumoto I, Kakuatani M, Yagi C, Morimoto M, Miyoshi A, et al. (2018) Spontaneous remission of panhypopituitarism with DI due to lymphocytic hypophysitis in a pregnant woman with pre-existing DI. Progr 220th Kinki Dist Meet Japanese Soc Intern Med 40: Presentation 35 (In Japanese).
• 67   Caturegli  P,  Newschaffer  C,  Olivi  A,  Pomper  MG,  Burger  PC, et al. (2005) Autoimmune hypophysitis. Endocr Rev 26: 599–614.
• 68   Iwama  S,  Sugimura  Y,  Kiyota  A,  Kato  T,  Enomoto  A, et al. (2015) Rabphilin-3A as a targeted autoantigen in lymphocytic infundibulo-neurohypophysitis. J Clin Endocrinol Metab 100: E946–E954.
• 69   Sakurai  K,  Yamashita  R,  Niituma  S,  Iwama  S,  Sugimura  Y, et al. (2017) Usefulness of anti-rabphilin-3A antibodies for diagnosing central diabetes insipidus in the third trimester of pregnancy. Endocr J 64: 645–650.
• 70   Suzukawa  M,  Takeda  K,  Suehiro  Y,  Tanizawa  Y (2015) A case of acute fatty liver of pregnancy complicated with diabetes insipidus. Nihon Naika Gakkai Zasshi (The J Japanese Soc Intern Med) 104: 2407–2411 (In Japanese).
• 71   Sheehan  HL (1940) The pathology of acute yellow atrophy and delayed chloroform poisoning. BJOG Int J Obstet Gynaecol 47: 49–62.
• 72   Vigil-De Gracia  P (2001) Acute fatty liver and HELLP syndrome: two distinct pregnancy disorders. Int J Gynecol Obstet 73: 215–220.
• 73   Zhang  YP,  Kong  WQ,  Zhou  SP,  Gong  YH,  Zhou  R (2016) Acute fatty liver of pregnancy: a retrospective analysis of 56 cases. Chin Med J (Engl) 129: 1208–1214.
• 74   Goel  A,  Jamwal  KD,  Ramachandran  A,  Balasubramanian  KA,  Eapen  CE (2014) Pregnancy-related liver disorders. J Clin Exp Hepatol 4: 151–162.
• 75   Goel  A,  Ramakrishna  B,  Zachariah  U,  Ramachandran  J,  Eapen  CE, et al. (2011) How accurate are the Swansea criteria to diagnose acute fatty liver of pregnancy in predicting hepatic microvesicular steatosis? Gut 60: 138–139.
• 76   Ch’ng  CL,  Morgan  M,  Hainsworth  I,  Kingham  JGC (2002) Prospective study of liver dysfunction in pregnancy in Southwest Wales. Gut 51: 876–880.
• 77   Yamanaka  Y,  Takeuchi  K,  Konda  E,  Samoto  T,  Satou  A, et al. (2002) Transient postpartum diabetes insipidus in twin pregnancy associated with HELLP syndrome. J Perinat Med 30: 273–275.
• 78   Konoshita  T,  Kuroda  M,  Kawane  T,  Koni  I,  Miyamori  I, et al. (2004) Treatment of congenital nephrogenic diabetes insipidus with hydrochlorothiazide and amiloride in an adult patient. Horm Res 61: 63–67.
• 79   Hatano  T,  Ogawa  K,  Kanda  K,  Seo  H,  Matsui  N (1988) Effect of ovarian steroids on cyclic adenosine 3':5'-monophosphate production stimulated by arginine vasopressin in rat renal monolayer cultured cells. Endocrinol Jpn 35: 267–274.
• 80   Okada  O,  Taguchi  A,  Kaneko  M,  Sugino  N (2006) A case of nephrogenic diabetes insipidus discovered in the wake of pregnancy. Chugoku Shikoku Dist J Japanese Obstet Gynecol Soc 54: 277 (In Japanese).
• 81   Nakao  Y,  Inde  Y,  Hayashi  M,  Takimoto  K,  Ozaki  K, et al. (2012) A case of nephrogenic diabetes insipidus diagnosed during pregnancy. Tokyo J Obstet Gynecol 61: 265–269 (In Japanese).
• 82   Kondo  T,  Nakamura  M,  Kawashima  J,  Matsumura  T,  Ohba  T, et al. (2019) Hyperemesis gravidarum followed by refeeding syndrome causes electrolyte abnormalities induced rhabdomyolysis and diabetes insipidus. Endocr J 66: 253–258.
• 83   Kwashima  A,  Matsuda  H,  Yosida  M,  Hasegwa  Y,  Yoshinaga  Y, et al. (2010) A case of renal diabetes insipidus due to lithium carbonate associated with pregnancy. Kanto J Obstet Gynecol 47: 226 (In Japanese).
• 84   Mizuno  O (1987) Transient nephrogenic diabetes insipidus associated with acute hepatic failure in pregnancy. Endocrinol Jpn 34: 449–455.
• 85   Oe  N (1988) A case of hereditary nephrogenic diabetes insipidus showing 47,XXX chromosomes and exacerbation of late pregnancy symptoms. J Japanese Soc Intern Med 77: 1113 (In Japanese).
• 86   Masuzaki  H,  Ichikawa  K,  Oe  N (1986) A case of pregnancy with nephrogenic diabetes insipidus. Perinat Med 16: 1511–1516 (In Japanese).
• 87   Teruya  S,  Nakasone  S,  Raita  Y,  Shimoji  K,  Miyazato  H, et al. (2009) A case of acute renal failure after delivery. Okinawa Med J 47: 151 (In Japanese).
• 88   Ala  Y,  Morin  D,  Mouillac  B,  Sabatier  N,  Vargas  R, et al. (1998) Functional studies of twelve mutant V2 vasopressin receptors related to nephrogenic diabetes insipidus: Molecular basis of a mild clinical phenotype. J Am Soc Nephrol 9: 1861–1872.
• 89   Guyon  C,  Lussier  Y,  Bissonnette  P,  Leduc-Nadeau  A,  Lonergan  M, et al. (2009) Characterization of D150E and G196D aquaporin-2 mutations responsible for nephrogenic diabetes insipidus: importance of a mild phenotype. Am J Physiol Renal Physiol 297: F489–F498.
• 90   Demura  M,  Takeda  Y,  Yoneda  T,  Oda  N,  Bando  Y, et al. (2004) Surgical stress-induced transient nephrogenic diabetes insipidus (NDI) associated with decreased Vasopressin receptor2 (AVPR2) expression linked to nonsense-mediated mRNA decay and incomplete skewed X-inactivation in a female patient with a heterozygous AVPR2 mutation (c. 89-90 delAC). Clin Endocrinol (Oxf) 60: 773–775.
• 91   Keller  SR,  Davis  AC,  Clairmont  KB (2002) Mice deficient in the insulin-regulated membrane aminopeptidase show substantial decreases in glucose transporter GLUT4 levels but maintain normal glucose homeostasis. J Biol Chem 277: 17677–17686.
• 92   Keller  SR (2004) Role of the insulin-regulated aminopeptidase IRAP in insulin action and diabetes. Biol Pharm Bull 27: 761–764.