Volume Management With Peritoneal Dialysis

Physical Exam

Frequent and routine volume assessment among patients receiving PD remains a cornerstone of providing high-quality care. Physical exam maneuvers can yield clues to the volume status of the patient; however, volume assessment remains imprecise and suffers from interobserver variability. These maneuvers include BP assessment (postural if needed), weight, peripheral edema, rales on lung auscultation, third heart sound on cardiac evaluation, and jugular venous pressure (JVP). None of these features alone is enough to determine volume status, and multiple signs and investigations together should be used to determine the volume status of patients.

Assessment of BP is a key parameter for PD patients, as volume overload underlies an important mechanism for hypertension in dialysis patients. In hemodialysis (HD) patients, Dry-weight Reduction in Hypertensive Hemodialysis Patients (DRIP): A Randomized, Controlled Trial demonstrated that a reduction in dry weight was associated with improvements in BP control among hypertensive HD patients. In PD patients, similar findings were demonstrated by Gunal et al. in a small single-center study of 46 PD patients with hypertension, in whom they discontinued BP medications and studied the effects of sodium restriction and enhanced ultrafiltration (UF) (using 4.25% [3.86% anhydrous] solution) on BP; they found that with strict sodium restriction and aggressive UF, there was a significant decrease in BP, a decline in patient weight, and decreased cardiothoracic index (a marker of cardiomegaly, measured on chest x-ray). There is paucity of literature examining the ideal target BP in PD patients; thus, BP goals are largely extrapolated from general population data. As per the Kidney Disease Improving Global Outcomes (KDIGO) Hypertension in CKD Guidelines, both diabetic and nondiabetic patients with albuminuria greater than 30 mg/24 hours should target BP < 130/80 mm Hg, while those without albuminuria (less than 30 mg/24 hours) should target BP < 140/90 mm Hg regardless of diabetic status. In the recent KDIGO Controversies Conference on Hypertension and Volume Management in Dialysis, many of the difficulties with BP management in those receiving kidney replacement therapy were reviewed. Particularly, it is unclear what BP targets should be employed for patients receiving dialysis with cardiovascular disease as they have been largely excluded from many studies yet constitute a large segment of HD and PD populations. A recent large observational cohort study of 7335 incident PD patients in the Zhejiang Renal Data System in China evaluated the average BP over the first 3 months of therapy and its association to all-cause mortality and cardiovascular mortality. They found a U-shaped curve, with both hypertension (systolic BP > 141 mm Hg, diastolic BP > 85 mm Hg, mean arterial pressure [MAP] > 102 mm Hg, and pulse pressure > 60 mm Hg) and hypotension (systolic BP < 119 mm Hg, diastolic BP < 67 mm Hg, MAP < 88 mm Hg) being associated with higher all-cause mortality. While this study provides insights on the association with early BP abnormalities in PD patients, it did not encompass longitudinal changes in BP and antihypertensive therapy that may have modified the association between BP and mortality. Even the appropriate method to measure BP is unclear for PD patients; in the HD population, it has been identified that the best means to measure BP is home 44-hour ambulatory monitoring. If ambulatory BP monitoring is unavailable due to resource limitation or adherence by the patient, home BP measurements would be the next preferred modality, followed by in-office measurements, and finally, predialysis and postdialysis measurements being the least useful. There have been no studies assessing out-of-unit BP monitoring in PD patients. Given the paucity of data, it can be extrapolated from HD literature that in those receiving PD, ambulatory or home BP measurements would be the preferred method and, by virtue of the autonomous nature of the therapy, should be readily available as part of routine clinical care and aided by the emergence of remote patient monitoring technologies.

Weight should be routinely followed, as excessive weight gain on PD can be a sign of fluid retention and has been associated with worse outcomes. It is important to follow weight longitudinally, as a single measurement is not helpful, but serial values can yield insight into daily variations in volume status. There are inaccuracies that can occur with weight if not measured appropriately. Patients should weigh themselves on the same scale at the same time of day with minimal clothing to avoid inaccuracies in weight measurement. In the established PD patient, daily weights can be a useful tool to assess and manage volume status.

Additional methods of volume assessment can be used to supplement the physical examination. BIA, lung water, and inferior vena cava (IVC) ultrasound (US) are all tools that can help one further delineate the volume status of patients, and biomarkers including B-type naturetic peptide (BNP) and N-terminal fragment of the precursor proBNP (NT-proBNP) can also be used to evaluate volume status.

Bioelectric Impedance Analysis

BIA is used to assess the total body water, extracellular body water, and intracellular body water of a patient; it is a safe, economical, and portable means of assessment. Using low-voltage electrical currents determines the fluid composition of body compartments and thus has been of interest as a means to better assess volume status. The technique, however, has its limitations, including lower accuracy in women, pregnant patients, and those with pacemakers, and is affected by significant obesity, protein energy wasting, and prior limb amputations. Multiple groups have investigated the prognostic significance of BIA derangements in dialysis patients (including the IPOD-PD and EuroBCM studies described earlier), and generally, it is accepted that BIA provides a reasonably accurate assessment of hydration and lean body mass; however, care in interpretation must be taken as multiple factors can affect this measurement. Such factors include the nutritional health of patients, as protein-energy wasting results in a loss of fat and muscle and a relative overhydration on BIA. It is unclear if correcting this relative overhydration in those with protein-energy wasting is of clinical benefit, and it has not been established if correction would lead to deleterious effects on RKF. Furthermore, BIA does measure extracellular fluid volume but does not provide a measure of intravascular volume, which may pose a challenge to distinguish among those patients who may have extracellular fluid volume expansion yet normal or reduced intravascular volume versus patients with increases in both extracellular and intravascular volume. There have been very limited studies including PD patients and, within these studies, there is heterogeneity of patients with respect to levels of RKF. In a systematic review, the use of BIA as part of routine clinical care has not been shown to improve mortality in PD patients, although it does lead to higher rates of extracellular water decline when used routinely in follow-up. One would then postulate if extracellular water decline occurs that there would be a positive effect on BP control with use of BIA; however, the data at this time are conflicting. A single-center randomized control trial (RCT) of 240 patients by Tian et al. compared volume status management guided by traditional clinical exam methods versus BIA and showed improvement in fluid management over the short-term with BIA, but no significant improvements in BP, mortality, or technique survival. However, a prior systematic review by Covic et al. that included seven RCTs encompassing 1312 patients similarly found that BIA did not have significant improvement in mortality for PD patients, but the use of BIA did improve systolic BP control. Very few of these studies used mortality as an outcome for BIA and were underpowered to achieve realistic and modest differences in mortality risks. The Effect of Bioimpedance and Vitamin D versus Usual care on left ventricular mass in PD patients (FLUID study) is a multicenter RCT across sites in Canada that is evaluating the role of BIA-guided fluid management versus standard of care on left ventricular mass as measured by cardiac magnetic resonance imaging (MRI), the results of which should be imminently forthcoming. Currently, while BIA may have some clinical value in assessing hydration status, its utility for reduction in mortality and BP is uncertain in PD patients. BIA should be used with caution and only as an adjunct to the clinical exam to establish a volume status and may be helpful to quantify volume removal targets for the patient, but its role remains controversial given the findings presented earlier.

Lung Water Ultrasound

Lung water evaluation for extravascular water has been well studied and is an established method to evaluate for pulmonary congestion in patients on dialysis. Studies of lung US in PD patients show that a high percentage of patients (46%) have lung congestion. US of the lungs should include multiple windows in different planes (midclavicular, anterior axillary, midaxillary). Confirmation of three or more B-Lines (white “comet tails” arising from the pleural line continuing to the bottom of the screen) in two or more bilateral lung zones is considered pathological and suggests increased fluid in the lungs. Supportive data exist in the HD population for the use of lung US in fluid management; however, there is a paucity of such data in PD patients. A recent RCT by Loutradis et al. showed that compared to standard clinical assessment, use of lung US-guided UF resulted in a reduction in BP and greater dry weight reduction in HD patients. It is important to remember that other pathological findings, such as lung fibrosis or consolidation, will also produce B-Lines, and these results need to be interpreted in the clinical context of the patient. Furthermore, B-Lines only assess for extracellular lung water and do not provide an assessment of intravascular volume status, which limits diagnostic utility in patients with fluid compartment disorders.

Inferior Vena Cava Ultrasound

IVC measurements include the diameter of the IVC expressed as an index of body surface area (IVC index, units mm/m ² ) and collapsibility index (maximal diameter on expiration minus minimal diameter on deep inspiration) divided by maximal diameter on expiration, with diameters measured at 3 cm from the right atrium. IVC measurements can be used to assess for both volume overload and fluid responsiveness (as defined as an increase in stroke volume of 10%–15% after a 500 mL bolus of crystalloid over 10–15 minutes, indicating possible volume depletion) and have been shown as a useful tool in assessment of dry weight for HD patients. Generally, an IVC collapsibility index of less than 50% indicates fluid responsiveness, whereas an index greater than 75% indicates volume overload. There are limited data for the use of IVC measurements in PD patients; however, Toprak et al. assessed IVC measurement in 69 continuous ambulatory PD (CAPD) patients and found that it was independently predictive of left ventricular geometry; specifically, volume overload is known to be associated with left ventricular dilation and eccentric left ventricular hypertrophy, and IVC dilation on US was seen to be predictive of eccentric left ventricular hypertrophy geometry. Of course, there are multiple drawbacks to IVC measurement, most notably that it is a measurement of increased right atrial pressures and is not directly a measurement of volume status; thus, increased IVC index and collapsibility index can be seen in any condition that increases right atrial pressures (e.g., heart failure, pulmonary hypertension) or any condition that leads to the development of tricuspid regurgitation (e.g., pulmonary hypertension, heart failure, or coronary artery disease). Finally, the IVC is susceptible to anything that causes changes in intrathoracic pressure, and there is significant interuser variability in IVC measurements; thus, this technique should only be used by experienced operators and interpreted in the context of the patient’s other comorbidities.

Biochemical Markers of Volume Overload

BNP is primarily secreted by the ventricular myocardium but can also be expressed from other tissues, including the brain, adrenal glands, kidneys, and lungs. Transcription of BNP occurs under pathophysiologic stressors, including stretch or strain of the myocardium occurring due to pressure or volume overload, as well as other factors including ischemia, proinflammatory cytokines, sympathetic overactivity, and vasoactive peptides. The physiologic roles of BNP include inducing diuresis, peripheral vasodilatation, and inhibition of the renin-angiotensin-aldosterone system (RAAS) and the sympathetic nervous system. The NT-proBNP is often now used in place of BNP measurement due to its longer half-life (120 minutes compared with 20 minutes for BNP). Both BNP and NT-proBNP are found to be elevated in patients with reduced kidney function, and this is felt to be secondary to both increased production due to the high frequency of ventricular dysfunction and left ventricular hypertrophy in the dialysis population and decreased clearance of the peptide. Thus, a single measurement in PD patients is often not of significant diagnostic utility. While it is well established that both BNP and NT-proBNP are cleared by HD, there is a paucity of literature regarding the clearance of PD. One small study of 11 PD patients by Obineche et al. examined serum BNP and NT-proBNP levels present in PD patients before, during, and after nightly intermittent PD (NIPD) treatment and found that PD had no effect on the concentration of these peptides in the serum. However, its utility may be a serial marker, as changes in BNP and NT-proBNP levels correlate with volume status; increases correlate with worsening volume overload and reduction of BNP and NT-proBNP, as associated with improvement in overhydration measurements in the PD-specific bioimpedance literature. Ultimately, longitudinal BNP and NT-proBNP measurements in PD patients may be helpful to assess changes in volume status, but interpretation is complicated by a number of factors, including known higher baseline values in dialysis patients, multiple stimuli for secretion in dialysis patients, unclear knowledge of the role of peritoneal clearance, and the impact of cardiovascular comorbidities on their interpretation.

Left Ventricular Mass

Left ventricular mass increases or hypertrophy is commonly seen in dialysis patients and present in up to 74% of incident dialysis patients. Multiple factors contribute to the development of left ventricular hypertrophy, including anemia, hypertension, volume overload, and disorders of bone mineral metabolism. There are multiple different ways to assess for left ventricular hypertrophy, including electrocardiogram (ECG), echocardiography, and MRI. Electrical diagnosis with ECG has largely been replaced with anatomical measures such as MRI and echocardiogram due to the low sensitivity of ECG for detection of left ventricular hypertrophy. An echocardiogram is generally considered the gold standard for diagnosis of left ventricular hypertrophy in hypertensive disease; however, the 2017 American Heart Association Hypertension guidelines do not discuss preference of echocardiography over MRI, and MRI has been shown to be an efficacious and useful tool for identifying left ventricular hypertrophy in centers that have access to this imaging modality. The advantages of MRI include being more reproducible than echocardiogram, better assessment of diastolic dysfunction and atrial volumes, and does not rely on geometric assumptions like echocardiography does. However, the pitfalls include a long exam time; limited availability and expertise, particularly in resource-poor settings; high cost; and an inability to perform the exam in patients with metallic implants. Echocardiography, while widely available, has limitations when evaluating left ventricular hypertrophy, including high interobserver and interstudy variability and assessment of diastolic dysfunction (one of the first complications of hypertension) changes with heart rate and preload variability, significantly affecting measurements. Ultimately, given the lack of availability for MRI, generally, echocardiography is the investigation of choice for left ventricular hypertrophy, and MRI is reserved for those cases where there is discrepancy between ECG and echocardiogram, severely progressive left ventricular hypertrophy, inconclusive echocardiogram, or in evaluation of hypertrophic cardiomyopathy. Left ventricular hypertrophy has been used as a surrogate for cardiovascular outcomes in patients with kidney disease, and as such, reductions in left ventricular hypertrophy are an endpoint for many trials in patients with chronic kidney disease (CKD). This use stemmed from the Heart Outcomes Prevention Evaluation (HOPE) and Losartan Intervention For Endpoint reduction in cardiovascular morbidity and mortality (LIFE) trials, which established that interventions that lead to regression of left ventricular hypertrophy are associated with lower cardiovascular mortality, myocardial infarction, stroke, and heart failure in patients with essential hypertension. However, in a systematic review and meta-analysis by Badve et al., 73 trials that used the change in left ventricular hypertrophy as an outcome found that there is no consistent correlation between intervention-induced reduction in left ventricular hypertrophy and mortality in CKD patients. Taken together, whether or not left ventricular hypertrophy is a direct consequence of fluid overload in PD patients or a therapeutic target for reduction in extracellular fluid volume remains unclear.

Volume status assessment remains a complex issue for patients on dialysis, and despite advances with US, BIA, and left ventricular hypertrophy, subjective clinical assessment still remains the standard of care for volume assessment in patients receiving kidney replacement therapy, and use of these additional tools has not been shown to affect mortality. See Fig. 24.1 for a summative graphic description of volume assessment in PD patients.