**1.1 Definition and epidemiology of Acute Kidney Injury (AKI) and obstetric AKI**

Insight into the occurrence and consequences of kidney disease has rapidly progressed. More than 30 different definitions have been used for defining AKI in the literature, creating much confusion and making comparisons difficult (Bellomo et al, 2001). Recently, consensus

<sup>\*</sup> Corresponding Author

Epidemiology, Causes and Outcome of Obstetric Acute Kidney Injury 69

The incidence of obstetric AKI has further declined over last 4 decades (Prakash et al, 2010). This improvement is due to improved availability of safe and legal abortion, more widespread and aggressive antibiotic use decreasing the incidence of post-abortal sepsis, and improved prenatal care. In the past, obstetric AKI used to be mostly due to post-abortal sepsis (Gul et al, 2004). Four decade long retrospective review of pregnancy related AKI cases was published from a centre in Italy. They reported virtual non-existence of post abortal sepsis, while AKI from obstetric complications like amniotic fluid embolism, extensive haemorrhage and prolonged intrauterine death etc. had decreased. AKI associated with preeclampsia and eclampsia remained stable until 1987 however its incidence decreased dramatically thereafter contributing towards the improved maternal mortality rate. The decreased occurrence was more evident in developed nations (Stratta et al, 1996). In 1995, the maternal mortality ratio in Africa was estimated to be over 1000 per 100 000 pregnancies and in Europe 28 per 100 000 pregnancies (Hill et al., 2001). In a recent review on AKI amongst all hospital admissions, pregnancy related AKI has not been listed as one of the causes of AKI. Authors have included critically ill patients with AKI in this review and have reviewed around 170 published literatures (Waikar et al, 2008). It may not be an over

However in recent reports published from developing countries; the frequency of obstetric AKI have been reported to be varying between 4-15% (Sivakumar et al, 2011). In a single centre study from India the incidence of AKI was reported as 1 in 56 births (Prakash et al, 2010) in contrast to 1 in 20000 births as reported from Italy (Stratta et al, Ren Fail 1996). Similar studies from individual centre from various developing countries have been published over the last decade emphasizing the fact that obstetric AKI is still prevalent in the developing or poor income nations (Table 2). However due to the absence of metaanalysis on obstetric AKI especially from these countries it is difficult to precisely estimate

statement to say that these cases are declining in incidence furthermore.

Fig. 1. Incidence of pregnancy related AKI.

its actual incidence.

definitions and classification systems have been proposed for AKI. RIFLE criteria stratify AKI into five stages (Table 1). This will eventually allow consistency across studies such that results can be compared (Ricci et al, 2011).


Small but important differences are observed between the two systems. A time constraint of 48 h for diagnosis (using either serum creatinine levels or urine output) is required in AKIN criteria. GFR decreases are used for diagnosis only in RIFLE criteria. In both systems, only one criterion (creatinine or urine output) has to be met to qualify for a given class or stage of AKI. Classes L and E of the RIFLE criteria are not reported. Owing to the wide variation in indications for and timing of initiation of RRT, individuals who receive RRT are considered to have AKIN Stage 3 AKI irrespective of their serum creatinine level and urine output.6,15 Abbreviations: AKI, acute kidney injury; AKIN, AKI Network; GFR, glomerular filtration rate; RIFLE, Risk, Injury Failure, Loss, End-stage renal disease; RRT, renal replacement therapy.

Table 1. Classification and staging systems for AKI.

AKI is estimated to occur in as many as 4%-20% of hospital admissions (Waikar et al, 2008) and in approximately 5-6% of critically ill patients with the period prevalence ranging from 1-25% (Uchino et al, 2005). Similarly, the period prevalence for acute renal replacement therapy in ICU is around 4-5% (Uchino et al, 2005). Septic shock in itself attributes to 50-60% of the cases. Hospital mortality in critically ill patients with AKI is equally higher at approximately 60% and has been reported to range from 28-90% (Uchino et al, 2005; Bellomo et al, 2004).

Incidence of pregnancy related AKI used to be 24-40% of all AKI in 60's which decreased to 2-3% in 80's (Fig 1) (Stratta, 1996). Interestingly enough, its incidence was already decreasing in 1963 as compared to 1959 when AKI occurred in 1 in 5000 pregnancies and 1 in 1400 pregnancies respectively in developed countries (Smith et al, 1965; Knapp & Hellman, 1959).

definitions and classification systems have been proposed for AKI. RIFLE criteria stratify AKI into five stages (Table 1). This will eventually allow consistency across studies such that

System Serum creatinine criteria Urine output criteria

2 Serum creatinine increase >2.0–3.0-fold from baseline <0.5 ml/kg/h for 12 h

Small but important differences are observed between the two systems. A time constraint of 48 h for diagnosis (using either serum creatinine levels or urine output) is required in AKIN criteria. GFR decreases are used for diagnosis only in RIFLE criteria. In both systems, only one criterion (creatinine or urine output) has to be met to qualify for a given class or stage of AKI. Classes L and E of the RIFLE criteria are not reported. Owing to the wide variation in indications for and timing of initiation of RRT, individuals who receive RRT are considered to have AKIN Stage 3 AKI irrespective of their serum creatinine level and urine output.6,15 Abbreviations: AKI, acute kidney injury; AKIN, AKI Network; GFR, glomerular filtration

rate; RIFLE, Risk, Injury Failure, Loss, End-stage renal disease; RRT, renal replacement therapy.

AKI is estimated to occur in as many as 4%-20% of hospital admissions (Waikar et al, 2008) and in approximately 5-6% of critically ill patients with the period prevalence ranging from 1-25% (Uchino et al, 2005). Similarly, the period prevalence for acute renal replacement therapy in ICU is around 4-5% (Uchino et al, 2005). Septic shock in itself attributes to 50-60% of the cases. Hospital mortality in critically ill patients with AKI is equally higher at approximately 60% and has been reported to range from 28-90% (Uchino et al, 2005;

Incidence of pregnancy related AKI used to be 24-40% of all AKI in 60's which decreased to 2-3% in 80's (Fig 1) (Stratta, 1996). Interestingly enough, its incidence was already decreasing in 1963 as compared to 1959 when AKI occurred in 1 in 5000 pregnancies and 1 in 1400 pregnancies respectively in developed countries (Smith et al, 1965; Knapp & Hellman, 1959).

<0.5 ml/kg/h for 6 h

<0.5 ml/kg/h for 12 h

<0.5 ml/kg/h for 6 h

<0.3 ml/kg/h for 24 h OR anuria for 12 h OR

need for RRT

Anuria for 12 h

Risk Serum creatinine increase to 1.5-fold OR GFR decrease

Injury Serum creatinine increase to 2.0-fold OR GFR decrease

Failure Serum creatinine increase to 3.0-fold OR GFR decrease

1 Serum creatinine increase ≥26.5 μmol/l (≥0.3 mg/dl) OR increase to 1.5–2.0-fold from baseline

3 Serum creatinine increase >3.0-fold from baseline OR

>75% from baseline OR serum creatinine ≥354 μmol/l (≥4 mg/dl) with an acute increase of at least 44

serum creatinine ≥354 μmol/l (≥4.0 mg/dl) with an acute increase of at least 44 μmol/l (0.5 mg/dl) OR

results can be compared (Ricci et al, 2011).

>25% from baseline

>50% from baseline

μmol/l (0.5 mg/dl)

need for RRT

Table 1. Classification and staging systems for AKI.

RIFLE class

AKIN Stage

Bellomo et al, 2004).

The incidence of obstetric AKI has further declined over last 4 decades (Prakash et al, 2010). This improvement is due to improved availability of safe and legal abortion, more widespread and aggressive antibiotic use decreasing the incidence of post-abortal sepsis, and improved prenatal care. In the past, obstetric AKI used to be mostly due to post-abortal sepsis (Gul et al, 2004). Four decade long retrospective review of pregnancy related AKI cases was published from a centre in Italy. They reported virtual non-existence of post abortal sepsis, while AKI from obstetric complications like amniotic fluid embolism, extensive haemorrhage and prolonged intrauterine death etc. had decreased. AKI associated with preeclampsia and eclampsia remained stable until 1987 however its incidence decreased dramatically thereafter contributing towards the improved maternal mortality rate. The decreased occurrence was more evident in developed nations (Stratta et al, 1996). In 1995, the maternal mortality ratio in Africa was estimated to be over 1000 per 100 000 pregnancies and in Europe 28 per 100 000 pregnancies (Hill et al., 2001). In a recent review on AKI amongst all hospital admissions, pregnancy related AKI has not been listed as one of the causes of AKI. Authors have included critically ill patients with AKI in this review and have reviewed around 170 published literatures (Waikar et al, 2008). It may not be an over statement to say that these cases are declining in incidence furthermore.

Fig. 1. Incidence of pregnancy related AKI.

However in recent reports published from developing countries; the frequency of obstetric AKI have been reported to be varying between 4-15% (Sivakumar et al, 2011). In a single centre study from India the incidence of AKI was reported as 1 in 56 births (Prakash et al, 2010) in contrast to 1 in 20000 births as reported from Italy (Stratta et al, Ren Fail 1996). Similar studies from individual centre from various developing countries have been published over the last decade emphasizing the fact that obstetric AKI is still prevalent in the developing or poor income nations (Table 2). However due to the absence of metaanalysis on obstetric AKI especially from these countries it is difficult to precisely estimate its actual incidence.

Epidemiology, Causes and Outcome of Obstetric Acute Kidney Injury 71

that may show a decline in the GFR include chronic glomerular diseases, lupus nephritis, diabetic nephropathy, and the chronic interstitial nephritides. Discussion on renal function during pregnancy in women with underlying renal disease is beyond the scope of this chapter. Finally, there are causes of renal failure specific to pregnancy, which are relatively more commonly encountered in gravid women. AKI seems to have bimodal distribution during

pregnancy in first trimester and third trimester respectively.

**Acute renal Failure in During early pregnancy** 

**• ATN resulting from a septic abortion**  ARF Late in Pregnancy/Postpartum •Thrombotic microangiopathy

•Renal Cortical Necrosis resulting from

–ATN from septicaemia or hypotension

–Mild-moderate hydronephrosis is normal

–Nephrolithiasis if solitary functioning kidney Table 3. Causes of Acute kidney Injury in Pregnancy.

–Occasionally, degree of obstruction sufficient to cause ARF

Prerenal azotemia is the most common cause of both community and hospital-acquired AKI. It is an appropriate physiological response to renal hypo perfusion (Blantz, 1998). Similarly, the most common cause of AKI specifically in first trimester of pregnancy is prerenal azotemia owing to hyperemesis gravidarum or vomiting from acute pyelonephritis. In late pregnancy volume contraction may be secondary to blood loss (Pertuiset & Grunfled, 1994). Antepartum and postpartum haemorrhage have been reported as major causes of AKI in pregnancy (Smith et al, 1965; Kennedy et al, 1973; Ali et al, 2004). Pregnant women who sustained AKI receive blood transfusion frequently; this further emphasizes the frequency of significant haemorrhage in these patients (Ali et al, 2004; Khanal et al, 2010). Uterine bleeding secondary to abortion or septic abortions are still chief causes of obstetric AKI in

•Intrinsic renal disease/autoimmune diseases

–Prolonged intrauterine foetal death

–Amniotic fluid embolism

–ARF from micro abscesses •Acute fatty liver of pregnancy

•Acute pyelonephritis

Obstructive Causes

developing countries.

–Severe preeclampsia usually with HELLP syndrome

**•Acute or massive blood loss** 

**–Abortion –Ectopic** 

–TTP-HUS

–Placenta previa

**–Hyaditiform mole •Severe dehydration –Ac. Pyelonephritis –Hyperemesis gravidarum** 


**\*** Study population included all pregnant women with AKI, causes listed existed together or individually **\*\***only included patients in third trimester; NA: not applicable; this series had sepsis as cause of AKI in 16 (18.8%) patients

Table 2. Incidence and causes of Obstetric AKI in developing countries.

### **1.2 Causes of AKI (Table 3)**

First of all, any cause which can give rise to AKI in non-pregnant women of reproductive age group (pre-renal, renal, and post renal or obstructive) may contribute to AKI in pregnancy. They can be coincidental simultaneous occurrences in pregnant women. As in any evaluation of AKI, causes not related to pregnancy must thus be considered and excluded. Pregnancy unrelated causes of AKI represent only about 5% of all obstetric AKI (Krane, 1988).

A second group of patients are those with underlying chronic renal disease that worsens during pregnancy, suggesting the development of acute renal failure superimposed on their chronic diseases. Glomerular diseases might be diagnosed for the first time in pregnancy and AKI may occur as a result of rapidly progressive glomerulonephritis. Examples of disorders

**APH/ PPH** 

1995 426 13.9 45 2 4 - 5 3

2006 190 - 130 11 33 - 16 -

2008 70 9.06 14 27 20 23 43 17

2008 40 7.02 20 8 6 4 - 6

2010 50 - 3 36 14 - 16 18

2010 4758 1.78 NA 16 30 14 21 18

2010 75 - 11 9 57 - - -

2011 59 4.36 28 11 18 - - 11

2010 106 20.75 Only included patients with preeclampsia

**Pre Eclampsia/ HELLP** 

**DIC Puerperal** 

**sepsis Misc.** 

**Septic abortion**

Chugh et al 1976 325 22.1 31 12 8 9 9 12

Kilari et al 2006 41 4.24 4 7 10 - 12 8

Hassan et al 2009 130 33 - 24 5 4 12 - \*Khalil et al 2009 60 - 3 22 10 11 1 4

Agida et al 2010 46 13 Only included patients with preeclampsia

**\*** Study population included all pregnant women with AKI, causes listed existed together or individually **\*\***only included patients in third trimester; NA: not applicable; this series had sepsis as

First of all, any cause which can give rise to AKI in non-pregnant women of reproductive age group (pre-renal, renal, and post renal or obstructive) may contribute to AKI in pregnancy. They can be coincidental simultaneous occurrences in pregnant women. As in any evaluation of AKI, causes not related to pregnancy must thus be considered and excluded. Pregnancy unrelated causes of AKI represent only about 5% of all obstetric AKI

A second group of patients are those with underlying chronic renal disease that worsens during pregnancy, suggesting the development of acute renal failure superimposed on their chronic diseases. Glomerular diseases might be diagnosed for the first time in pregnancy and AKI may occur as a result of rapidly progressive glomerulonephritis. Examples of disorders

Table 2. Incidence and causes of Obstetric AKI in developing countries.

**Author Year** 

Prakash et al

\*Prakash et al

Goplani et al

Saleem Najar et al

\*Khanal et al

Prakash et al

\*\*Prakash et al

\*Erdemoglu et al

Sivakumar et al

(Krane, 1988).

cause of AKI in 16 (18.8%) patients

**1.2 Causes of AKI (Table 3)** 

**published**

**Study popn** 

**Obs-AKI (%)** 

that may show a decline in the GFR include chronic glomerular diseases, lupus nephritis, diabetic nephropathy, and the chronic interstitial nephritides. Discussion on renal function during pregnancy in women with underlying renal disease is beyond the scope of this chapter.

Finally, there are causes of renal failure specific to pregnancy, which are relatively more commonly encountered in gravid women. AKI seems to have bimodal distribution during pregnancy in first trimester and third trimester respectively.


Table 3. Causes of Acute kidney Injury in Pregnancy.

Prerenal azotemia is the most common cause of both community and hospital-acquired AKI. It is an appropriate physiological response to renal hypo perfusion (Blantz, 1998). Similarly, the most common cause of AKI specifically in first trimester of pregnancy is prerenal azotemia owing to hyperemesis gravidarum or vomiting from acute pyelonephritis. In late pregnancy volume contraction may be secondary to blood loss (Pertuiset & Grunfled, 1994). Antepartum and postpartum haemorrhage have been reported as major causes of AKI in pregnancy (Smith et al, 1965; Kennedy et al, 1973; Ali et al, 2004). Pregnant women who sustained AKI receive blood transfusion frequently; this further emphasizes the frequency of significant haemorrhage in these patients (Ali et al, 2004; Khanal et al, 2010). Uterine bleeding secondary to abortion or septic abortions are still chief causes of obstetric AKI in developing countries.

Epidemiology, Causes and Outcome of Obstetric Acute Kidney Injury 73

with severe preeclampsia, usually with the HELLP syndrome (haemolysis with a microangiopathic blood smear, elevated liver enzymes, and a low platelet count) (McCrae et

There are anecdotal reports on sporadic cases of acute glomerulonephritis (GN) including post infectious GN, good-pasture's syndrome, lymphoma, drug nephrotoxicity, incompatible blood transfusions and endocarditis causing AKI during pregnancy (Pertuiset & Grunfeld, 1994). Acute interstitial nephritis (from antibiotics or non-steroidal antiinflammatory drugs etc.) may coincidentally occur during pregnancy and result in AKI.

To understand the pathophysiology and proper management of renal problems in pregnancy it is important that we are familiar with the anatomical and physiological

The kidneys enlarge during normal pregnancy, increasing by 1 to 2 cm in length and in volume by up to 70% towards term, due to tissue hypertrophy and expansion of both interstitial and vascular compartments. More important from a clinical perspective is the increase in size of the renal pelvices and ureters. By third trimester about 80% of the pregnant women have hydronephrosis which is easily evident by ultrasound, more on the right than on the left (Baylis & Davison , 2010; Brown et al, 2010). A number of factors are thought to be important in this change. Progesterone, a smooth muscle relaxant, reduces ureteric tone and peristalsis. The asymmetric dilation of the pelvicalyceal system suggests extrinsic compression by the enlarging uterus at the pelvic brim, hypertrophy of surrounding connective tissue (Waldeyer's sheath) and kinking due to ligaments or compression by iliac blood vessels (Brown et al, 2010). The clinical consequence of these changes can be urinary stasis; increasing the risk of bacterial growth and asymptomatic bacteriuria of pregnancy. If the changes are in extreme and precipitate the over distention syndrome, with massive dilation they may present with symptoms like recurrent severe flank pain, increasing serum creatinine, hypertension, or even reversible acute kidney injury (Khanna & Nguyen, 2001). In case of presentation with over distension symptoms that suggest renal colic but no stone detectable by ultrasound or by radiographic imaging; it is imperative to exclude urinary tract infection and to avoid the temptation to drain the system using nephrostomy tubes. Furthermore it is important to remember that acute renal failure as a consequence of ureteric obstruction in pregnancy is uncommon; and that ureteric dilation is part of normal pregnancy and it is not usually possible to distinguish between

Detailed discussion of systemic haemodynamics of normal pregnancy is beyond the scope of this chapter. Briefly, changes start as early as the first trimester with reduced systemic vascular resistance, increase in cardiac output by 40-50% and resting tachycardia by 24th week (Davison & Dunlop, 1980). There is progressive expansion of the plasma and

**1.3 Changes in renal anatomy and physiology in normal pregnancy:** 

al, 1992) both can cause AKI.

**1.3.1 Renal tract anatomy** 

changes that occur during normal pregnancy.

this and pathologic dilation (Brown et al, 2010).

**1.3.2 Renal physiology** 

**1.3.2.1 Systemic haemodynamics** 

Pregnancy-specific conditions such as preeclampsia, HELLP syndrome, acute fatty liver of pregnancy (AFLP), haemolytic uremic syndrome/ thrombotic thrombocytopenic purpura independently or in combination cause uterine bleeding ante-partum or post-partum haemorrhage. These conditions are frequently associated with complications like abruptioplacentae, hepatic infarction, hepatic rupture, intra-abdominal bleeding, and puerperal sepsis all of which can be further complicated by AKI. This occurs frequently in third trimester (Krane, 1988; Maynard et al, 2007; Prakash, 2010). Preeclampsia is a multi-system disorder unique to human pregnancy characterised by hypertension and involvement of one or more other organ systems and/or the foetus. American College of Obstetrics and Gynaecology, requires blood pressures >140/90 mm Hg on two occasions combined with urinary protein excretion >300 mg/d for the diagnosis of preeclampsia. Preeclampsia occurs in 3-5% of pregnancies and is associated with increased maternal and fetal mortality especially in developing countries. It is a leading cause of premature deliveries in developed countries thus increases the neonatal morbidity (Society of Obstetric Medicine of Australia and New Zealand [SOMANZ], 2009).

Abruptio placentae and puerperal sepsis; may also occur independent of these conditions and can be complicated by AKI. Obstetric complications such as septic abortion and placental abruption are associated with severe acute tubular necrosis (ATN) and bilateral cortical necrosis. Acute cortical necrosis, usually involves bilateral renal cortex, may occur as a consequence of irreversible or severe ATN. It has been found to be associated with poor renal outcome in longer term. Bilateral cortical necrosis is most often a complication of abruptio-placentae (36 per cent in the series of Chugh); while in other studies it was associated with disseminated intravascular coagulation. It presents as acute renal failure in other conditions too but, unlike acute tubular necrosis, total and persistent anuria is almost constant (Kleinknecht et al, 1973; Chugh, 1976). The diagnosis can be established either by renal biopsy or, better, by selective renal angiography. Other imaging studies (plain radiograph, ultrasound scan, CT-scan of abdomen and nuclear renal scan) may also be helpful. Renal cortical necrosis which occurs as a consequence of AKI in pregnancy continues sporadically (Naqvi et al, 1996; Prakash et al, 2007). With overall decrease in the incidence of AKI in pregnancy and improved overall management; incidence of cortical necrosis is decreasing even in the developing countries. Whenever present it is associated with irreversible renal failure (Khanal, 2010). Sepsis is still a major cause including septic abortions and puerperal sepsis in several studies published from India over last decades (Sivakumar, 2011).

Less common and miscellaneous causes of obstetric AKI include:

Obstructive uropathy; obstruction may occur in gravidas due to polyhydramnios, incarcerated gravid uterus, or can occur even in women with otherwise uncomplicated gestation due to retroverted uterus. Rarely, acute urinary tract obstruction in pregnancy is induced by a kidney stone, and it seldom causes renal failure (Strothers & Lee, 1992; Scarpa et al, 1996).

Amniotic fluid embolism which occurs primarily in multi-para after prolonged labour can cause AKI. Those with underlying renal parenchymal disease even without advanced chronic kidney disease are more prone to develop acute tubular necrosis (ATN) especially due to super imposed pre-eclampsia (Pertuiset & Grunfeld, 1994). Thrombotic thrombocytopenic purpura-haemolytic uremic syndrome (TTP-HUS) can easily be confused There are anecdotal reports on sporadic cases of acute glomerulonephritis (GN) including post infectious GN, good-pasture's syndrome, lymphoma, drug nephrotoxicity, incompatible blood transfusions and endocarditis causing AKI during pregnancy (Pertuiset & Grunfeld, 1994). Acute interstitial nephritis (from antibiotics or non-steroidal antiinflammatory drugs etc.) may coincidentally occur during pregnancy and result in AKI.
