RENAL DATA FROM ASIA–AFRICA
|Year : 2020 | Volume
| Issue : 2 | Page : 493-502
|The changing face of pregnancy-related acute kidney injury from eastern part of India: A hospital-based, prospective, observational study
Sandeep Saini1, Arpita Ray Chaudhury1, Smita Divyaveer1, Pratima Maurya2, Dipankar Sircar1, Sanjay Dasgupta1, Debabrata Sen1, Sambhunath Bandyopadhyay3, Rajendra Pandey1
1 Department of Nephrology, Institute of Post-Graduate Medical Education and Research and Seth Sukhlal Karnani Memorial Hospital, Kolkata, West Bengal, India
2 Department of Obstetrics and Gynaecology, North 24 Parganas District Hospital, Barasat, West Bengal, India
3 Department of Obstetrics and Gynaecology, Institute of Post-Graduate Medical Education and Research and Seth Sukhlal Karnani Memorial Hospital, Kolkata, West Bengal, India
Click here for correspondence address and email
|Date of Submission||10-Jan-2019|
|Date of Decision||06-Mar-2019|
|Date of Acceptance||09-Mar-2019|
|Date of Web Publication||09-May-2020|
| Abstract|| |
This study was initiated to look into the etiologies, prevalence, and outcome of pregnancy-related acute kidney injury (PRAKI) in a tertiary care hospital. Women admitted with PRAKI from January 2015 to December 2016 were included in the study. All patients were investigated and treated and followed up for the next six months.. For statistical analysis, Chi- square test and analysis of variance were performed to analyze the data. Multivariate analysis was applied to compare the risk of nonrecovery of renal function in different etiologies of PRAKI. During the study period, 81 patients were admitted with PRAKI, of whom 68 (84%) received hemodialysis (HD). A total of 449 patients including all cases of AKI underwent HD from January 2015 to June 2016. The incidence of dialysis requiring PRAKI was 68 out of the 449 patients (15%). Sixty-eight (84%) patients required dialysis support while the most common cause was sepsis (49%), with the second being pregnancy-associated atypical hemolytic-uremic syndrome (P-aHUS) (17%) followed by obstetric hemorrhages (16%). There was a significant reduction of first-trimester AKI (8.6%) compared to a previous study published from this institute (19.3%). The maternal mortality (25%) and fetal mortality (23.5%) were high. Nearly 39% of the patients had complete recovery of renal function. This study revealed significant PRAKI burden due to a largely preventable cause, puerperal sepsis. Renal survival was poor in P- aHUS. The gaps in the obstetric care may be identified for the improvement of fetomaternal outcome.
|How to cite this article:|
Saini S, Chaudhury AR, Divyaveer S, Maurya P, Sircar D, Dasgupta S, Sen D, Bandyopadhyay S, Pandey R. The changing face of pregnancy-related acute kidney injury from eastern part of India: A hospital-based, prospective, observational study. Saudi J Kidney Dis Transpl 2020;31:493-502
|How to cite this URL:|
Saini S, Chaudhury AR, Divyaveer S, Maurya P, Sircar D, Dasgupta S, Sen D, Bandyopadhyay S, Pandey R. The changing face of pregnancy-related acute kidney injury from eastern part of India: A hospital-based, prospective, observational study. Saudi J Kidney Dis Transpl [serial online] 2020 [cited 2021 Feb 25];31:493-502. Available from: https://www.sjkdt.org/text.asp?2020/31/2/493/284025
| Introduction|| |
Pregnancy-related acute kidney injury (PRAKI) has been one of the most challenging riddles faced by nephrologists as obstetric AKI has been associated with poor outcome in comparison to other causes of AKI, affecting both fetal and maternal well-being. In India, though the incidence has reportedly come down over the last few decades, still, it constitutes a significant proportion of dialysis requiring AKI and is often multifactorial. The incidence of PRAKI in the developed countries is only 1%-2%, whereas in the developing countries, it is about 4.2%-15%,, with a different etio- logical spectrum and of these, 25% of PRAKI require referral to dialysis center. In low- resource settings, it still represents 20% of all cases of AKI, associated with a high incidence of fetal/ neonatal (39%) and maternal (20%) mortality. Maternal mortality, in turn, is an indicator of population health and socioeco-nomic development. Studies from various parts of India have reported different causes and outcomes, possibly reflecting the variability of the study popu-lation in geo-ethnic aspects, education and awareness, and peripheral health-care accessibility. A previous study from our institute published in the late last decade highlighted high maternal mortality rate (MMR) (28.1%), with cause-specific mortality being highest with sepsis (36.8%) and hemorrhage (25%).
Hence, the study was planned to register an institution (a tertiary referral teaching institute of eastern part of India)-based incidence etiology and outcome of PRAKI along with an attempt to identify factors associated with poor prognosis and to analyze the time trends with particular focus on factors amenable for intervention.
| Materials and Methods|| |
A hospital-based, prospective, observational study was conducted from January 2015 to December 2016 at Post-Graduate Medical Education and Research and Seth Sukhlal Karnani Memorial Hospital, Kolkata. The study was approved by the institutional ethical committee. Patients who satisfied the criteria of AKI as defined by the Kidney Disease Improving Global Outcomes (KDIGO) were recruited after written consent. Patients who had prior renal disease or small-sized kidney on ultrasonography were excluded from the study.
Patients’ old records and treatment history including medical, surgical, and obstetric history were recorded. All patients underwent investigations and treatment as per protocol which has been given as supplementary data. The need for renal replacement therapy was based on patients’ symptoms and clinical status in accordance with the recommendations by the KDIGO. Renal biopsy was done in case of nonrecovering of renal function at the end of three weeks or in anuric patients as earliest.
Outcomes recorded were maternal death due to any cause, fetal death, if any, and renal outcomes, defined as complete recovery if estimated glomerular filtration rate (eGFR) was more than 60 mL/min/1.73 m2 or non- recovery if eGFR was <60 mL/min/1.73 m2 persisting for three months. eGFR was calcu-lated by the Modification of Diet in Renal Disease method. All patients were followed up with laboratory investigations at four-week interval till six months after discharge and after six months at three-monthly interval.
| Statistical Analysis|| |
Descriptive statistical analysis was performed to calculate the means with corresponding standard deviations. Chi-square test was performed to find the associations. One-way analysis of variance followed by post hoc Tukey’s test was performed with the help of critical difference or least significant difference at 5% and 1% level of significance to compare the mean values. t-test was used to compare the means. P <0.05 was taken to be statistically significant. Multivariate analysis was applied to compare the risk of nonrecovery of renal function in different etiologies of PRAKI.
| Results|| |
During the study period, 81 patients were diagnosed with PRAKI at our institution, of whom 68 required dialysis. A total of 449 patients were admitted with dialysis requiring AKI. The institutional incidence of dialysis requiring PRAKI was 68 out of the 449 patients (15%). Twelve out of the 81 (15%) patients underwent delivery in our institute, while the rest 69 out of the 81 patients (85%) had delivery in outside hospitals and were then subsequently referred to our institute. During this study period, 5645 obstetric cases were admitted for delivery in our institute, of which 12 patients developed PRAKI (i.e., one out of 470 pregnant women admitted in our institute developed AKI).
The baseline characteristics and demographic profile of the patients are summarized in [Table 1]. The number of patients having PRAKI in the first, second, and third trimesters including post-partum were four (4.9%), five (6.17%), and 72 (88.88%), respectively.
The mean age of the patients was 23.64 ± 4.59 years, ranging from 14 to 38 years, and the median age was 22.0 years. Patients in the age group of 20-24 years (40.7%) were statistically significantly higher than those of other age groups, i.e., 14 to 19 years, 25 to 29 years, 30 to 34 years, and 35 to 39 years (Z = 2.41; P = 0.016). The median number of pregnancies for each woman was two (range 1-5). The median period of gestation of the patients was 37 weeks (range 8-41 weeks). Thirty-four (42%) of all the PRAKI patients had normal vaginal delivery, while 39 (48%) patients had lower uterine cesarean section (LUCS). The average serum creatinine at presentation was 5.65 mg/dL.
The spectrum of the etiology of PRAKI is shown in [Table 2]. As shown in [Table 2], majority of the patients, i.e., 40 (49%), presented with features of sepsis alone such as fever; pelvic pain; and blood-stained, foul- smelling, vaginal discharge along with leuko- cytosis (no other contributing factor to AKI). Sepsis occurred post LUCS in 18 out of the 40 patients. Of the 40 patients with sepsis AKI, 34 (85%) were referred from primary care centers, and culture reports were available for only four patients who were positive for Escherichia coli. Six patients who delivered at our center had sepsis, two of them grew Staphylococcus aureus, and the rest were positive for Group B Streptococci. All the four patients in the first-trimester PRAKI had medical termination of pregnancy (MTP)-related sepsis. These MTPs were elective. In the second trimester, one of the two patients had illegal abortion and in other, MTP was done in view of rapidly declining renal function.
The second most common etiology was biopsy-proven thrombotic microangiopathy (TMA) in 14 out of the 81 (17.28%) patients. Five out of the 14 received plasmapheresis in view of the clinical possibility of atypical hemolytic-uremic syndrome (aHUS), who at their clinical presentation had some features of sepsis (93% of all TMAs) such as leuko- cytosis, but the presence of blood transfusion requiring anemia in the absence of significant blood loss, thrombocytopenia, normal coagulation profile raised lactate dehydrogenase, and schistocytes in peripheral blood smear helped to suggest pregnancy-associated-aHUS (P- aHUS). The rest of the eight patients had histopathological features of TMA in renal biopsy. One patient who was clinically suspected to have P-aHUS died before renal biopsy.
Ten patients presented with pregnancy- induced hypertension (PIH), of whom four had features of hemolysis, elevated liver enzymes and thrombocytopenia syndrome and renal dysfunction. Out of the 10 patients with PIH and PRAKI, four required dialysis support; all the four had leukocytosis and three of these had simultaneous Gram-negative septicemia as well.
Twenty-four (30%) patients underwent renal biopsy. The histopathological spectrum of these PRAKI patients [Table 3] shows that TMA with acute cortical necrosis (ACN) is the most common cause of nonrecovery of renal function followed by ACN due to sepsis and obstetric hemorrhage.
|Table 3: Histopathological spectrum of pregnancy-related acute kidney injury.|
Click here to view
Sixty-eight out of the 81 (84%) patients received hemodialysis support [Table 2]. Before initiation of dialysis, one patient died due to sepsis with multiorgan failure.
Twenty out of the 81 (25%) patients died. The mortality in each of the etiological group is shown in Table 2. Maternal mortality is highest in sepsis patients. The diagnosis of acute fatty liver of pregnancy was suggested in view of presentation in the third trimester, absence of hypertension, negative viral sero- logy for hepatitis viruses, elevated trans- aminases with rise of bilirubin, and presence of coagulopathy along with the absence of any biliary tract pathology on USG. All deaths were recorded within four weeks of admission.
Excluding lost to follow-up (11 patients), the renal outcome of the rest (70 patients) is shown in [Figure 1]. Complete recovery of renal function was seen in 27 out of the 70 patients (39%), while 23 out of the 70 patients (33%) had nonrecovery of renal function. At six months, 11 out of the 70 patients (16%) had dialysis-dependent renal failure.
|Figure 1: Renal outcome of pregnancy-related acute kidney injury at 6 months.|
Click here to view
On multivariate analysis, it was found that the risk for renal nonrecovery was 14 times more for TMA and the risk was statistically significant [OR - 14.00 (1.13, 172.58); P = 0.039], followed by the risk of nonrecovery for obstetric hemorrhage [OR - 7.50 (0.62, 90.61);
P = 0.11] and sepsis [OR - 4.28 (0.44, 41.35); P = 0.20], but, later, the two risk ratios were not significant. Of all the clinical and laboratory parameters studied, multivariate analysis showed that oliguria, anemia, and hypo- albuminemia were significantly associated with nonrecovery of renal function.
Fetal outcome was relatively poor, with fetal death occurring in 19 (23.5%) cases overall. Of these fetal deaths, 11 out of the 19 were associated with maternal sepsis.
| Discussion|| |
This study has revealed that PRAKI in the eastern part of India is still a major cause of dialysis requiring AKI. While most of the previous studies have looked into only dialysis requiring renal failure, this study registers nondialysis requiring AKIs as well. Eighty- four percent of the patients were requiring dialysis and most of them (85.2%) were referred from other peripheral health-care centers as this hospital serves as a tertiary referral center for a large catchment area.
The demographic profile of this study is similar to that of other studies except a significantly high incidence of PRAKI in early age group of 14-19 years, which is consistent with the National Health Survey report published in 2016, revealing high prevalence of early age marriage (27% in urban and 46% in rural areas) in West Bengal [Table 4]. The incidence of PRAKI is relatively high in rural area where antenatal coverage is still poor (38%). As per the National Family Health Survey conducted in West Bengal which is also consistent with our data, 37% of pregnant women in 1992 and 25% in 2015 to 2016 did not receive antenatal care., Although it is difficult to establish the direct causal relationship, poor antenatal visits are associated with increased incidence of PRAKI. Despite the improved number of institutional deliveries, puerperal sepsis remains the highest contributor to PRAKI, identifying poor infection control practices in rural and peripheral healthcare delivery centers.
Chugh et al reported that 59% of women developed AKI after abortion in the first trimester in 1975. When compared with the studies of PRAKI published in India in the last 10 years, this study along with other studies, reveals a decrease in the incidence of early- trimester (i.e., <20 weeks’ gestation) PRAKI (8.6% in the present study), which used to be mostly due to septic abortion, along with a rise in late-trimester PRAKI (91.4%) due to puerperal sepsis, severe pre-eclampsia, P-aHUS, and antepartum hemorrhage. After legalization of abortion, more public awareness about abortion-related complications of illegal abortion, and availability of better reproductive health-care facilities especially through the National Rural Health Mission, the reported incidence of abortion-related PRAKI has reduced to 55%-8% in the last decade.
Sepsis following delivery or abortion is the most common contributor to PRAKI as well as maternal mortality, similar to the previous reports (75%), from other parts of India. Few studies published after 2010 have documented decrease in puerperal sepsis, but it was still the major cause of PRAKI in our study. An alarming finding was the high incidence of sepsis in spite of the increase in the number of institutional deliveries. Most of these septic patients developed dialysis requiring AKI and eventually had high case fatality rate due to multiorgan failure. The previous study from our institute has also highlighted the issue.
P-aHUS has emerged as the second highest cause of PRAKI after sepsis. P-aHUS as an etiology of PRAKI has been reported quite rarely, i.e., around 1% to 3%, in the previous decade. In one Indian study by Prakash et al where they looked at a composite outcome of three decades, very low incidence of TMA (0.06%) was found. The occurrence of leuko- cytosis in TMA closely mimics sepsis, making it difficult to diagnose unless a high index of suspicion is maintained and specific noninvasive investigations are asked for, while deferring renal biopsy if needed until the general condition of patient improves. While P-aHUS may be triggered by sepsis, ante- partum hemorrhage (APH), and postpartum hemorrhage (PPH) as well, the reported incidence of complement-mediated gene dysregu- lation is also high in these P-aHUS patients. In a published French study, P-aHUS occurred mostly in second pregnancy and 79% presented in puerperium. Complement abnormalities were detected in 18 out of 21 patients. In our study, 93% of patients presented in puerperium and 28% progressed to nonrecovery of renal function.
Obstetric hemorrhage (APH and PPH) appeared as the third most common cause of PRAKI (16%) in the present study. It comprised around 20%-40% in previous studies.,, PPH occurring postcesarean, complicated with sepsis, raised the case fatality rate, again highlighting the need for vigilance in peripheral health-care facilities.
This study showed that 12% of patients had preeclampsia or eclampsia at presentation, while our institute reported high incidence as 26% in 2007-2009, similar to studies conducted in Varanasi (38%), Chennai (21%), Kashmir (15%), and Pakistan (28%). This may reflect that an early diagnosis and prompt treatment of preeclamptic toxemia has resulted in lesser incidence of AKI in this group. However, international reports reveal an excess of preeclampsia or eclampsia as cause of PRAKI because of pregnancy in older age when hypertension and diabetes are becoming common comorbidities with conspicuous absence of sepsis-associated PRAKI. The frequency of preeclampsia ranges between 2% and 7% in healthy nulliparous women and increases substantially in women with multiple gestation, chronic hypertension, diabetes mellitus, and chronic kidney disease (CKD).
Anemia was clinically present in 91% of patients, again indicating an inadequacy of antenatal care. Anemia increases the risk of developing AKI by decreasing the threshold of AKI in cases of antepartum hemorrhage. A study from Karnataka shows that 100% of 24 patients were anemic at presentation and in Kashmir, 80% of patients were found to be anemic during the study period.
Leukocytosis and hyperbilirubinemia were seen in higher proportions in our study as compared to other studies from India, probably related to the higher percentage of sepsis in our study. Hypoproteinemia and hypoalbu- minemia were significantly high, as documented by other studies as well, and associated with poorer outcome. Oliguria was present in 85% of patients and may be due to late referral from peripheral hospitals.
During hospital stay, 84% of the patients required dialysis support. In terms of complete recovery of renal function, our study has found a particularly worse outcome (16% are dialysis dependent). Most of the patients had late referral and reached our institute with advanced renal failure and septic shock, complicated by the presence of significant anemia and hypo- albuminemia which came out as significant prognostic markers of nonrecovery. The study by Gopalkrishnan et al also had looked into the prognostic markers and had a similar observation like us. Other studies from various parts of India revealed recovery (partial or complete) of renal function in 40%-70% of patients.,
In our study the maternal mortality was high in those who required dialysis (n = 20, 25%) . Other studies from India reported a mortality rate of 15%-30%., The WHO definition of MMR does not register maternal death after 42 days, but AKI per se as well as CKD resulting as a sequel of AKI does contribute to mortality in these patients beyond 42 days postpartum.
One more interesting aspect of this study is reporting biopsy finding in 24 (30%) patients out of the 81 patients. Among those who underwent a renal biopsy, TMA was found in 54% in addition to acute tubular necrosis and complete or patchy ACN. Most of these patients progressed to end-stage renal disease in six -month follow-up. CAN may result from sepsis, TMA, or APH or by various combinations of these three factors. The etiology of ACN cannot be established on biopsy alone. We have used the corroborative clinical evidence to categorize ACN further into likely etiologies. The biopsy proof of TMA may identify a group of population who will need screening for complement abnormalities. Nearly 41% of our biopsy cases had ACN due to TMA, sepsis, and obstetric hemorrhage. Similar high rates of ACN (30%) were reported in another study from Pakistan. Prakash et al had shown a decline in the incidence of PRAKI with very low incidence of ACN (2.2%) as they reported sepsis as the predominant cause of ACN.
Nineteen (23.5%) fetal deaths were recorded in this study highlighting the need for improved antenatal care and early referral to higher center where advanced care may be required.
| Conclusion|| |
PRAKI contributes significantly to total number of AKI cases admitted in tertiary care center and is multifactorial in many cases. Sepsis, a largely preventable entity, continues to be a major cause of AKI despite increase in the number of institutional deliveries and is associated with poor outcome. Early referral to higher center for patients developing compromised renal function, improvement of nutritional status and anemia correction through better antenatal care, and increasing sepsis prevention awareness in peripheral health-care facilities may help reduce burden of PRAKI. As P-aHUS may have features overlapping with sepsis and associated with very poor renal outcome, a high index of suspicion should be maintained especially in postpartum PRAKI to detect this entity early, which may change the course the disease.
| Ethics committee|| |
As per memo number Inst/IEC/2015/190, the Institutional Ethical Committee of IPGME&R, Kolkata, approved to conduct this study.
Conflict of interest: None to declare.
| References|| |
Gammill HS, Jeyabalan A. Acute renal failure in pregnancy. Crit Care Med 2005;33:S372-84
Chugh KS, Sitprija V, Jha V. Acute renal failure in tropical countries. In: Oxford Textbook of Clinical Nephrology. New York: Oxford University Press; 1998. p. 1714-34
Pertuiset N, Grunfeld JP. Acute renal failure in pregnancy. Baillieres Clin Obstet Gynaecol 1994;8:333-51
Prakash J, Niwas SS, Parekh A, et al. Acute kidney injury in late pregnancy in developing countries. Ren Fail 2010;32:309-13
Wilmoth JR, Mizoguchi N, Oestergaard MZ, et al. A new method for deriving global estimates of maternal mortality. Stat Politics Policy 2012;3:pii 2151-7509. 1038
Arora N, Mahajan K, Jana N, Taraphder A. Pregnancy-related acute renal failure in eastern India. Int J Gynaecol Obstet 2010;111:213-6
Kidney Disease Outcomes Quality Initiative. KDIGO clinical practice guidelines for acute kidney injury. Kidney Int Suppl 2012;2:1-38
Prakash J, Pant P, Prakash S, et al. Changing picture of acute kidney injury in pregnancy: Study of 259 cases over a period of 33 years. Indian J Nephrol 2016;26:262-7
Najar MS, Shah AR, Wani IA, et al. Pregnancy related acute kidney injury: A single center experience from the Kashmir Valley. Indian J Nephrol 2008;18:159-61
Krishna A, Singh R, Prasad N, et al. Maternal, fetal and renal outcomes of pregnancy- associated acute kidney injury requiring dialysis. Indian J Nephrol 2015;25:77-81
Gopalakrishnan N, Dhanapriya J, Muthukumar P, et al. Acute kidney injury in pregnancy - A single center experience. Ren Fail 2015;37: 1476-80
Aggarwal RS, Mishra VV, Jasani AF, Gumber M. Acute renal failure in pregnancy: Our experience. Saudi J Kidney Dis Transpl 2014;25: 450-5
Chugh KS, Singhal PC, Sharma BK, et al. Acute renal failure of obstetric origin. Obstet Gynecol 1976;48:642-6
Davison JM, Dunlop W. Renal hemodynamics and tubular function normal human pregnancy. Kidney Int 1980;18:152-61
Brenner and Rector’s The Kidney. 10th
ed. Philadelphia: Elsevier; 2016. p. 1176
Fakhouri F, Roumenina L, Provot F, et al. Pregnancy-associated hemolytic uremic syndrome revisited in the era of complement gene mutations. J Am Soc Nephrol 2010;21: 859-67
Prakash J, Pant P, Singh AK, Sriniwas S, Singh VP, Singh U. Renal cortical necrosis is a disappearing entity in obstetric acute kidney injury in developing countries: Our three decade of experience from India. Ren Fail 2015;37:1185-9
Dunlop W. Serial changes in renal haemo- dynamics during normal human pregnancy. Br J Obstet Gynaecol 1981;88:1-9
Prakash J. The kidney in pregnancy: A journey of three decades. Indian J Nephrol 2012;22: 159-67
Khanal N, Ahmed E, Akhtar F. Factors predicting the outcome of acute renal failure in pregnancy. J Coll Physicians Surg Pak 2010; 20:599-603
Mehrabadi A, Liu S, Bartholomew S, et al. Hypertensive disorders of pregnancy and the recent increase in obstetric acute renal failure in Canada: Population based retrospective cohort study. BMJ 2014;349:g4731
Dambal A, Lakshmi KS, Gorikhan G, et al. Obstetric acute kidney injury; a three year experience at a medical college hospital in north karnataka, India. J Clin Diagn Res 2015; 9:OC01-4
Ali A, Ali MA, Ali MU, Mohammad S. Hospital outcomes of obstetrical-related acute renal failure in a tertiary care teaching hospital. Ren Fail 2011;33:285-90.
Arpita Ray Chaudhury
Department of Nephrology, Institute of Post- Graduate Medical Education and Research and Seth Sukhlal Karnani Memorial Hospital, Kolkata, West Bengal
[Table 1], [Table 2], [Table 3], [Table 4]
| Article Access Statistics|
| Viewed||605 |
| Printed||11 |
| Emailed||0 |
| PDF Downloaded||85 |
| Comments ||[Add] |