Proteinuria in pregnancy: Diagnosis, differential diagnosis, and management of nephrotic syndrome

INTRODUCTION — Clinicians caring for pregnant people need to understand how to identify proteinuria in pregnancy and determine the cause. Proteinuria is one of the cardinal features of preeclampsia (table 1), a common and potentially severe complication of pregnancy. Proteinuria is also a sign of primary kidney disease or kidney disease secondary to systemic disorders, such as diabetes mellitus or hypertension. Adding to the complexity, 20 to 25 percent of pregnant people with chronic hypertension, diabetes mellitus, or chronic kidney disease develop superimposed preeclampsia [1,2].

This topic will discuss diagnosis of proteinuria in pregnancy, differential diagnosis, and management of pregnant people with nephrotic syndrome. The evaluation of proteinuria in nonpregnant individuals and measurement of protein excretion are discussed in detail separately. (See "Assessment of urinary protein excretion and evaluation of isolated non-nephrotic proteinuria in adults".)

Pregnancy issues in patients with kidney disease are also discussed separately:

•(See "Pregnancy and contraception in patients with nondialysis chronic kidney disease".)

•(See "Acute kidney injury in pregnancy".)

•(See "Pregnancy in patients on dialysis".)

DIAGNOSIS/DEFINITION — In pregnancy, urinary protein excretion normally increases substantially to 150 to 250 mg daily and is not considered abnormal until it exceeds 300 mg daily [3]. In many pregnant people, proteinuria appears to increase further during labor and delivery [4]. Abnormal proteinuria is similarly defined for singleton and multiple gestations, even though pregnant people with twin pregnancies have greater increases in urinary protein excretion than those with singleton pregnancies and may exceed 300 mg daily in the absence of any pathology [5,6]. By comparison, in nonpregnant individuals without kidney disease, normal urinary protein excretion is less than 150 mg daily and normal albumin excretion is than 30 mg daily [7].

Urinary protein excretion increases substantially during pregnancy due to a combination of increased glomerular filtration rate (GFR) and increased permeability of the glomerular basement membrane [8]. Additionally, tubular reabsorption of filtered protein is reduced in pregnancy, along with other nonelectrolytes, such as amino acids, glucose, and beta-microglobulin. Additional information on pregnancy-related changes in kidney function and the urinary tract can be found separately. (See "Maternal adaptations to pregnancy: Renal and urinary tract physiology".)

MEASUREMENT

Semiquantitative (screening) — The urinary dipstick is used to screen for proteinuria.

Standard urine dipstick testing should be performed on a fresh, clean voided, midstream urine specimen, obtained before pelvic examination to minimize the chance of contamination from vaginal secretions. The urinary dipstick for protein is a semi-quantitative colorimetric test that primarily detects albumin. Results range from negative to 4+, corresponding to the following estimates of protein excretion:

●Negative

●Trace – between 15 and 30 mg/dL

●1+ – between 30 and 100 mg/dL

●2+ – between 100 and 300 mg/dL

●3+ – between 300 and 1000 mg/dL

●4+ – >1000 mg/dL

A positive reaction (1+) for protein develops at the threshold concentration of 30 mg/dL, which crudely corresponds to a 24-hour urinary protein excretion of 300 mg/day, depending on urine volume.

Although inexpensive and commonly used, the urinary dipstick has a high false-positive and false-negative rate when used to screen for abnormal proteinuria in pregnancy, especially at the 1+ level [9,10]. This is due primarily to variability in urine concentration (osmolality), which can substantially affect random urine protein concentration (ie, the dipstick result).

False-positive tests may occur in the presence of gross (macroscopic) blood in the urine, semen, very alkaline urine (pH >7), quaternary ammonium compounds, detergents and disinfectants, drugs, radio-contrast agents, and concentrated urine (specific gravity >1.030). Positive tests for protein due to blood in the urine seldom exceed 1+ by dipstick.

False negatives may occur with dilute urine (specific gravity <1.010), high salt concentration, highly acidic urine, or nonalbumin proteinuria. False-negative urine dipstick proteinuria testing is most common in the third trimester of pregnancy, when the clinical findings of preeclampsia are most likely to manifest [11].

Because the urinary dipstick has low diagnostic accuracy for the detecting proteinuria in pregnancy, it is not recommended as a routine screening test for preeclampsia by the United States Preventive Services Task Force [12]. However, it is commonly performed and the detection of proteinuria at the level of 2+ or greater by urine dipstick may be used for the diagnosis of preeclampsia if quantitative methods are unavailable [13]. (See "Preeclampsia: Clinical features and diagnosis", section on 'Screening and risk reduction'.)

Quantitative (diagnostic) — Urinary protein can be measured as either albumin or total protein. There are three methods to quantify proteinuria: 24-hour urine collection, spot urine protein-to-creatinine ratio, and spot urine albumin-to-creatinine ratio.

24-hour urine collection — The traditional method to quantify proteinuria requires a 24-hour urine collection to directly measure the daily total protein or albumin excretion. If urine and serum creatinine are also measured, the 24-hour urine collection creatinine clearance can provide an estimate of the glomerular filtration rate.

The 24-hour collection is begun at the usual time the patient awakens. At that time, the first void is discarded and the exact time noted. Subsequently, all urine voids are collected with the last void timed to finish the collection at exactly the same time the next morning. The time of the final urine specimen should vary by no more than 5 or 10 minutes from the time of starting the collection the previous morning.

An inexpensive basin urinal that fits into the toilet bowl facilitates collection for pouring into a bottle. The bottle(s) may be kept at normal room temperature for a day or two, but should be kept cool or refrigerated for longer periods of time. No preservatives are needed. (See "Patient education: Collection of a 24-hour urine specimen (Beyond the Basics)".)

Although generally considered the "gold standard" for diagnosis of proteinuria in both preeclampsia and kidney disease, the 24-hour urine protein excretion in pregnant people is frequently inaccurate due to under-collection or over-collection [14]. Thus, when interpreting the results of a 24-hour urine collection, it is critical to assess the adequacy of collection by quantifying the 24-hour urine creatinine excretion, which depends on muscle mass. In a complete collection, the 24-hour urine creatinine excretion will be between 15 and 20 mg/kg body weight, calculated using prepregnancy weight. Values substantially above or below this estimate suggest over- and under-collection, respectively, and should call into question the accuracy of the 24-hour urine protein result.

In addition to the high rate of inaccurate/incomplete collection, the 24-hour urine sample is cumbersome for ambulatory patients, and the laboratory result is not available for at least 24 hours while the collection is being completed and analyzed [14].

Urine protein-to-creatinine ratio — The spot urine protein-to-creatinine ratio (UPCR) has become the preferred method for the quantification of proteinuria in the nonpregnant population because of its high accuracy, reproducibility, and convenience when compared with timed urine collection [15]. (See "Assessment of urinary protein excretion and evaluation of isolated non-nephrotic proteinuria in adults".)

For preeclampsia diagnosis, proteinuria is defined as a 24-hour urine protein excretion >300 mg, spot protein-to-creatinine ratio >0.3 mg/mg (>300 mg/g, or approximately >30 mg/mmol), or ≥2+ on urinary dipstick if quantitative measurement is unavailable [13].

Some laboratories and international guidelines use UPCR in units of mg protein per mmol creatinine (mg/mmol). To convert mg/mmol to mg/g, multiply by 8.84. An UpToDate calculator is available for calculating the UPCR using spot urine protein and spot urine creatinine values (calculator 1).

Most studies evaluating UPCR to quantify proteinuria in pregnancy were performed in patients with suspected preeclampsia. In a 2020 meta-analysis, the UPCR was highly correlated with the 24-hour urine protein measurement [16], as it is in nonpregnant adults. A urine protein-to-creatinine ratio less than 30 mg/mmol (260 mg/g) had a high sensitivity (91 percent) and specificity (89 percent) for the diagnosis of proteinuria >300 mg/day by 24-hour urine collection. Use of the UPCR has also been validated for baseline proteinuria quantification in early pregnancy [17].

However, many obstetricians still prefer use 24-hour urine collection to measure proteinuria in pregnancy.

Urine albumin-to-creatinine ratio — The urine albumin-to-creatinine ratio (UACR), like the UPCR, is measured using a random "spot" urine specimen. (See "Assessment of urinary protein excretion and evaluation of isolated non-nephrotic proteinuria in adults".)

The UACR can be performed using an automated analyzer, allowing immediate point-of-care testing that could be utilized in an antenatal clinic. Like the UPCR, the UACR (using a threshold between 20 and 60 mg albumin/g creatinine) is strongly predictive of significant proteinuria (>300 mg protein/day by 24-hour urine collection) in a high-risk obstetric population [18,19] and in patients with hypertensive pregnancies [20,21]. In one study, pregnant people with a spot UACR >312 mg albumin/g creatinine measured at 17 to 20 weeks of gestation were at almost eightfold higher risk of subsequently developing preeclampsia (relative risk [RR] 7.8) compared with those with UACR <312 mg albumin/g creatinine [22]. The UACR performs similarly to the UPCR with regard to prediction of adverse pregnancy outcomes [23].

Some laboratories report UACR in units of mg albumin per mmol creatinine (mg/mmol). To convert mg/mmol to mg/g, multiply by 8.84.

8- or 12-hour collection — Although not commonly used, measurement of protein in an 8-hour [24] or 12-hour [25] urine collection is a reasonable alternative to the 24-hour urine collection for quantification of proteinuria. In a systematic review including seven studies, >150 mg of protein in a 12-hour collection was highly predictive of >300 mg protein in a 24-hour collection (pooled sensitivity 92 percent [95% CI 86-96], specificity 99 percent [75-100]) [25]. Alternatively, the urine protein-to-creatinine ratio may be measured from an aliquot of urine taken from a 16- or 24-hour urine collection [26].

DIFFERENTIAL DIAGNOSIS OF PROTEINURIA — For patients in whom proteinuria is the prominent finding, distinguishing between preeclampsia and other forms of kidney disease can be challenging. Initial assessment should include determination of the gestational age at onset and quantity of proteinuria. If the onset is after 20 weeks gestation, preeclampsia is likely (table 1) (see "Preeclampsia: Clinical features and diagnosis"). If the onset is before 20 weeks gestation, primary or secondary kidney disease is likely (table 2).

Kidney disease versus preeclampsia — The distinction between kidney disease and preeclampsia is important because it affects prognosis and management. Primary or secondary kidney disease may require diagnostic testing (including kidney biopsy in some cases) and disease-specific therapy. In these patients, the usual aim is delivery at term. The definitive treatment for preeclampsia is delivery, which is typically performed preterm when severe features or fetal complications are present. (See "Pregnancy and contraception in patients with nondialysis chronic kidney disease" and "Preeclampsia: Antepartum management and timing of delivery".)

When proteinuria is documented before the 20^th week of gestation, underlying kidney disease is likely because preeclampsia rarely occurs before that time. Conversely, new-onset proteinuria after 20 weeks of gestation suggests preeclampsia. Approximately one-third of pregnant people who present with new proteinuria after 20 weeks of gestation eventually progress to preeclampsia [27,28], which may have severe features [29], and 25 percent of pregnant people with preeclampsia have proteinuria as their initial presenting sign [30]. In such cases, adverse pregnancy and neonatal outcomes are more common as compared with patients in whom hypertension was the first presenting sign of preeclampsia [30,31].

In preeclampsia, the degree of proteinuria is only weakly associated with adverse maternal and neonatal outcomes. Heavy proteinuria is not considered a severe feature of preeclampsia [32-36], although some studies report severe proteinuria (>3 to 5 g/day) is associated with earlier gestational age at preeclampsia onset, earlier gestational age at delivery, and a higher incidence of fetal growth restriction as compared with mild proteinuria [37,38]. Proteinuria may be absent in preeclampsia: up to 10 percent of patients with preeclampsia and 20 percent of those with eclampsia have no proteinuria at the time of initial presentation [39,40]. Proteinuria is no longer required for the diagnosis of preeclampsia, according to American College of Obstetricians and Gynecologists guidelines, if severe features of the disease are present (table 1) [13]. (See "Preeclampsia: Clinical features and diagnosis".)

Of course, de novo kidney disease (for example, lupus nephritis) can also occur later in pregnancy. Distinguishing between kidney disease and preeclampsia can be difficult in these cases, especially when information on protein excretion (and hypertension) in early pregnancy is unavailable. For this reason, it is useful to quantify protein excretion in early pregnancy in patients at risk for kidney disease (ie, patients with chronic hypertension, diabetes mellitus, and systemic lupus erythematosus).

The following algorithms and UpToDate topics are useful for evaluating patients suspected of having kidney disease (algorithm 1 and algorithm 2). Of note, urinary tract dilation (eg, >3 cm) on kidney ultrasound has been associated with proteinuria in pregnancy [41]. (See "Glomerular disease: Evaluation and differential diagnosis in adults" and "Chronic kidney disease (newly identified): Clinical presentation and diagnostic approach in adults".)

Measurement of soluble fms-like tyrosine kinase-1 (sFlt1) and placental growth factor (PlGF) may be helpful for the diagnosis of preeclampsia [42,43]. sFlt1 and PlGF are angiogenic factors secreted by the placenta and implicated in the pathogenesis of preeclampsia. The sFlt1:PlGF maternal serum assay is available in Europe and the United States. Although more studies are needed to validate its use, this serologic test may prove useful and cost-effective in distinguishing preeclampsia from other causes of proteinuria in pregnancy [44,45] and appears to be predictive of preeclampsia with severe features, adverse maternal outcomes, and adverse neonatal outcomes in patients with clinical suspicion of preeclampsia [46-48]. The sFlt1:PlGF ratio may be particularly helpful in distinguishing chronic kidney disease and preeclampsia [49]. (See "Preeclampsia: Pathogenesis", section on 'sFlt-1, VEGF, PlGF'.)

In some cases, the distinction between kidney disease and preeclampsia can only be made in retrospect, as clinical signs of preeclampsia generally resolve within 12 weeks after delivery [50], while proteinuria due to underlying kidney disease does not. However, resolution of proteinuria after preeclampsia, especially when severe features are present, can sometimes take much longer. In observational studies of patients with preeclampsia, it is common for proteinuria to persist for six months or longer after delivery [51]. Nevertheless, proteinuria (or hypertension) that persists longer than three months after delivery should prompt close follow-up and consideration of further evaluation and appropriate referral so that underlying renal disease or chronic hypertension is detected and treated expeditiously.

Worsening kidney disease versus superimposed preeclampsia — Preeclampsia is common in pregnant people with preexisting kidney disease and/or hypertension. Superimposed preeclampsia typically manifests as new or worsening hypertension, often with worsening proteinuria, after 20 weeks of gestation. A substantial (twofold or greater) increase in proteinuria may be a sign of impending preeclampsia [52]. Hematologic, neurologic, and hepatic complications (table 3) and/or evidence of fetal compromise (including fetal growth restriction) may also point to a diagnosis of superimposed preeclampsia.

Alternatively, worsening proteinuria in a pregnant patient with kidney disease may reflect the effect of pregnancy on the underlying kidney disease. Patients with chronic kidney disease often have a twofold or greater increase in proteinuria in pregnancy [53-55].

Nephrotic syndrome versus preeclampsia — Nephrotic-range proteinuria (>3 g/24 hours) is a sign of glomerular injury. Proteinuria less than 3 g/day is usually asymptomatic, while proteinuria >3 g/day may cause the nephrotic syndrome, with edema and hypoalbuminemia. The serum albumin concentration decreases in normal pregnancy, so it has been suggested that the nephrotic syndrome in pregnancy can be diagnosed by the presence of proteinuria >3 g/day and albumin below the lower limit of normal based on gestational age: <3.1 g/dL in the first trimester, <2.6 g/dL in the second trimester, and <2.3 g/dl in the third trimester [56,57].

Preeclampsia is the most common cause of de novo nephrotic-range proteinuria in pregnancy. In a retrospective study of 104 patients with nephrotic-range proteinuria during pregnancy who were evaluated postnatally, the diagnosis was preeclampsia alone in 60 percent, superimposed preeclampsia in 9 percent, newly diagnosed kidney disease in 9 percent, worsening proteinuria due to preexisting kidney disease in 6 percent (no hypertension), and isolated gestational proteinuria in 1 percent [58]. (See "Overview of heavy proteinuria and the nephrotic syndrome".)

Nephrotic range proteinuria during pregnancy due to primary glomerular disease is associated with a high risk of several adverse outcomes, including superimposed preeclampsia, acute kidney injury, preterm birth, low birth weight, and the need for neonatal intensive care [59]. Nevertheless, conservative management until delivery, particularly in patients who present in the third trimester, is often a reasonable management approach. When nephrotic syndrome presents early in pregnancy, and/or there is progressive decline in kidney function, timely treatment of the underlying kidney disease is often indicated. In such cases, the potential benefits of kidney biopsy may outweigh the risks. (See 'Role of kidney biopsy in pregnancy' below.)

ROLE OF KIDNEY BIOPSY IN PREGNANCY — In clinical practice, kidney biopsy is rarely needed in pregnancy. A clinical diagnosis is often established based on the history, physical examination, and serologic testing for primary and secondary glomerular disease. When indicated, serologic testing should be performed for lupus nephritis (ie, antinuclear antibodies and anti-double stranded DNA antibody titers and complement levels), membranous nephropathy (anti-PLA2R antibody), and infections associated with glomerular diseases (eg, HIV and hepatitis C virus). The decision to perform kidney biopsy during pregnancy is based on several factors, including the gestational age, the severity of the kidney disease, and the suspected underlying diagnosis.

Data on the safety of renal biopsy during pregnancy are limited. The major complication is bleeding. In a systematic review of reports of kidney biopsies performed during pregnancy or postpartum, the risk of bleeding was higher when the biopsy was performed during pregnancy as compared with postpartum (7 percent [16/197] versus 1 percent [3/268]) [60]. All observed cases of major bleeding (ie, requiring blood transfusion) occurred in biopsies performed between 23 to 26 weeks of gestation, suggesting patients in this gestational age range may be particularly vulnerable to complications. Early in pregnancy, the benefit of a biopsy (accurate diagnosis and subsequent tailored management) is more likely to outweigh its risks. Later in gestation, the gravid uterus makes the standard prone position for biopsy difficult. In such cases, the biopsy can be performed with the patient in the lateral decubitus position or deferred until the patient has stabilized postpartum.

Aspirin should be held for five to seven days prior to and one to two weeks following kidney biopsy to minimize the risk of bleeding. (See "The kidney biopsy".)

MANAGEMENT OF NEPHROTIC SYNDROME IN PREGNANCY — The management of nephrotic syndrome in pregnancy is based on expert opinion, as minimal data are available to support evidence-based practice. In nonpregnant patients with nephrotic syndrome, inhibitors of the renin-angiotensin aldosterone system (ie, angiotensin-converting enzyme inhibitors and angiotensin receptor antagonists) are routinely used as nonspecific therapy to reduce proteinuria. These agents are contraindicated in pregnancy, due to well-documented teratogenicity, particularly in the second and third trimesters of pregnancy. (See "Adverse effects of angiotensin converting enzyme inhibitors and receptor blockers in pregnancy".)

●Immunosuppressive drugs — First-line therapy for nephrotic syndrome caused by glomerular disease frequently includes immunosuppressive agents. Immunosuppressive drugs generally considered safe for use in pregnancy include azathioprine, glucocorticoids, and calcineurin inhibitors (eg, tacrolimus and cyclosporine). Mycophenolate mofetil and cyclophosphamide are teratogenic and should be avoided in pregnancy. Rituximab is generally avoided in pregnancy, though there are reports of its use without adverse fetal consequences [61,62]. There are inadequate data on the safety of belimumab, voclosporin, and anifrolumab in pregnancy, so these agents should also be avoided. (See "Safety of rheumatic disease medication use during pregnancy and lactation".)

●Low-dose aspirin — In pregnant patients with chronic kidney disease, low-dose aspirin starting before 16 weeks of gestation is recommended to reduce the risk of developing preeclampsia later in gestation. The use of aspirin for preeclampsia prevention is reviewed in detail separately. (See "Preeclampsia: Prevention" and "Preeclampsia: Prevention", section on 'Low-dose aspirin'.)

●Management of edema — Edema related to nephrotic syndrome in pregnancy can often be successfully managed without the use of diuretics. The dietary intake of sodium may be modestly limited. Graduated compression stockings and leg elevation are safe and often effective methods to facilitate resolution of edema.

Diuretics are avoided, if possible, because of the theoretical risk that they may impair pregnancy-associated expansion of plasma volume. However, there is no clear evidence of adverse fetal effects, and loop diuretics may be safely used in pregnant people with severe, refractory edema [63].

●Anticoagulation — Nephrotic syndrome is associated with an increased risk of deep venous thrombosis (DVT). Patients with severe nephrotic syndrome (particularly those with membranous nephropathy) and severe hypoalbuminemia who are not at high risk for bleeding may benefit from prophylactic anticoagulation (eg, with subcutaneous low molecular weight heparin). A suggested approach to prophylactic anticoagulation in pregnant people is presented in more detail elsewhere. (See "Hypercoagulability in nephrotic syndrome", section on 'Pregnant patients'.)

●Management of hyperlipidemia — Statins are avoided in pregnancy because of limited and contradictory data suggesting an increased risk of congenital anomalies with first-trimester exposure. We suggest discontinuing statins in females taking them because of nephrotic syndrome and planning pregnancy, and then resuming the drugs after delivery/breast feeding. (See "Statins: Actions, side effects, and administration", section on 'Risks in pregnancy and breastfeeding'.)

Bile acid sequestrants and fibrates have no established teratogenic effects and can be safely used in pregnancy to treat severe hyperlipidemia due to nephrotic syndrome.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Glomerular disease in adults" and "Society guideline links: Hypertensive disorders of pregnancy".)

INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5^th to 6^th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10^th to 12^th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.

Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)

●Beyond the Basics topic (see "Patient education: Protein in the urine (proteinuria) (Beyond the Basics)")

SUMMARY AND RECOMMENDATIONS

●Diagnosis – In pregnancy, urinary protein excretion normally increases substantially to 150 to 250 mg daily. (See 'Diagnosis/definition' above.)

●Screening – Urine dipstick is routinely used to screen for proteinuria in pregnancy, but false-positive and false-negative results are common. It is most predictive of abnormal proteinuria if +2 or greater. Positive urine dipsticks should be followed up with a quantitative test. (See 'Measurement' above.)

●Urine protein-to-creatinine ratio – The spot urine protein-to-creatinine ratio (UPCR; mg protein/mg creatinine) is an accurate, convenient, and relatively rapid method to quantify proteinuria in pregnancy. A UPCR greater than 0.3 mg/mg (30 mg/mmol) after 20 weeks gestation is one of the diagnostic criteria for preeclampsia (table 1). (See 'Quantitative (diagnostic)' above.)

●Differential diagnosis – The gestational age of proteinuria onset can help establish the diagnosis of preeclampsia versus kidney disease. (See 'Differential diagnosis of proteinuria' above.)

•Proteinuria documented before 20 weeks of gestation suggests preexisting kidney disease.

•Proteinuria beginning after 20 weeks gestation is usually due to preeclampsia (table 1), especially when associated with new or worsening hypertension or features of severe preeclampsia (table 3). If hypertension is absent, then primary or secondary kidney disease should be considered. Preeclampsia is the most common cause of de novo nephrotic-range proteinuria in pregnancy.

●Nephrotic syndrome

•In nephrotic syndrome, severe leg edema can be managed with sodium restriction, graduated compression stockings, and leg elevation. If edema persists despite these measures, loop diuretics may be used with caution. (See 'Management of nephrotic syndrome in pregnancy' above.)

•Nephrotic syndrome is associated with an increased risk of deep venous thrombosis (DVT). Patients with severe nephrotic syndrome and very low serum albumin levels who are not at high risk for bleeding may benefit from prophylactic anticoagulation. (See "Hypercoagulability in nephrotic syndrome", section on 'Pregnant patients'.)

•Bile acid sequestrants and fibrates can be safely used in pregnancy to treat severe hyperlipidemia due to nephrotic syndrome; statins should be avoided. (See 'Management of nephrotic syndrome in pregnancy' above.)