Hemodialysis: Daily and Nocturnal Dialysis

Table 1.

Organization	Endotoxin level (EU/ml)	Bacterial growth (cfu/ml)
Association for the Advancement of Medical Instrumentation (AAMI)	< 2	< 200
Canadian Standards Associations (CSA)	< 2	< 100
International Organization for Standardization (ISO)	< 0.25	< 100

Periodic water analysis needs to be done 2-3 times a year.

Water treatment for the NxStage machine

The NxStage system does not need water treatment if the dialysate is in bags. The PureFlow® system can produce ready dialysate.

Electricity and drainage

A dedicated electrical circuit is needed at home as well as an access to a drain.

Electricity

A dedicated GFI (Ground Fault Interrupter) breaker, wire and hospital grade receptacle for the dialysis machine must be installed from the main panel to the dialysis room.

The GFI amps and the wire size depend on the machine consumption (check manufacturer data sheet). An existing outlet is acceptable for the Water Reverse Osmosis machine

Water supply

The pressure of the incoming water must meet the WRO system requirements.

There must be both hot and cold water sources, usually taken from under the sink, connected to a blending valve and temperature gauge.

Where no city water exists, well water can be used to perform dialysis. Usually well water is hard and softeners are required. Chlorine is added to city water unlike well water and therefore a carbon filter is useful. The use of the carbon tanks is still useful also for well water as it is rich in iron removed by the carbon tanks.

Drain

The drain pipe is connected to the closest sink. It is important to pour bleach in the drain pipe on a monthly basis to prevent the buildup of protein inside the drain pipe. Water leaks may occur from the dialysis machine or the water treatment system.

Patients are advised to get plywood sheets completely covered with vinyl or a sheet of plastic. The dialysis machine will be placed over the covered plywood in case water leakage occurs from the hydraulic system of the machine. Water is kept within the frame parameters. Two water sensors should be placed on the plywood by the machine side where the water may leak.

Remote monitoring of the dialysis machine

Live remote monitoring of the dialysis machine during the night, through a telephone or internet connection, is employed by some dialysis centers. An observer calls the patient or the emergency services if machine alarms remain unattended. There is no consensus on the importance of such monitoring.

Remote monitoring is not considered mandatory in most jurisdictions. If it is available at a reasonable cost, it can provide reassurance to some patients during the initial period of nightly home dialysis and provides information about dialysis treatment compliance. Such information, however, can also be obtained through intermittent connection and interrogation of the dialysis machine.

Solute removal

Small molecules (urea)

The use of the weekly standard Kt/V (stdKt/V), as described by Gotch, has evolved into the preferred measure for dialysis dose for the alternative dialysis schedules. It is based on the predialysis mid-week urea levels. It therefore favors the frequent or continuous forms of dialysis (predialysis urea is lower). A stdKt/V of 2.3 represents the current KDOQI recommended dialysis dose (2015). The calculation of stdKt/V is more complex (14) than the use of the currently popular urea reduction ratio (URR).

Phosphate

Phosphate removal correlates well with the total length of dialysis per week. Therefore, SDHD for more than 2.5 hours daily and NHD are effective in terms of phosphate removal, resulting in fewer phosphate binders being required. No phosphate binders are needed for 6X/week NHD; neither for 70% of patients on every other night home NHD. Phosphate additives are needed in the dialysate of more than 50% of patients on NHD when done 6 nights a week. Successful removal of phosphate allows for liberal phosphate intake (see mineral metabolism below).

Large molecules

Removal of large molecular weight solutes depends on membrane flux, use of convection and most importantly on dialysis duration. The amount of beta-2 microglobulin removal on frequent NHD is four times as high as on conventional HD. This corresponds to the four times longer duration of NHD dialysis compared to conventional HD. Daily NHD and daily hemodiafiltration are associated with lower serum predialysis beta-2 microglobulin levels. Effects on highly protein bound molecules are minimal.

Urea kinetics in NHD

Formal urea kinetic modeling may not be accurate on daily and especially on NHD. Urea generation may not be constant on NHD due to the longer dialysis during a fasting state. There are also amino acid losses into the dialysate that can potentially affect urea generation. This may result in URR overestimation by 7-15%.

Dialysis access

AV fistulas, grafts and catheters have been used successfully for daily and NHD. AV fistulas are encouraged.

AV access

The buttonhole technique of fistula cannulation popularized by Twardowski has been widely used for home hemodialysis. It involves using the same site for successive cannulations. “Blunt” needles are utilized after the buttonhole track is established using sharp needles for 1 to 2 weeks. Before cannulation, the fistula and surrounding skin is cleansed with chlorhexidine gluconate (0.5% in 70% alcohol) or providone-iodine solution. Scabs are covered with an alcohol pad for 5 minutes and then removed using a sterile needle.

The advantages of this technique include diminished or no pain and the predictability of a successful cannulation. A high incidence of systemic infections has been encountered with the use of the buttonhole technique, mainly involving Staphylococcus aureus. The resulting bacteremias can lead to metastatic infections and significant morbidity, and even mortality. Meticulous aseptic techniques, especially the application of a small amount of mupirocin on the site after dialysis, have prevented this complication. Some have called for the abandonment of the buttonhole technique due to the risk of systemic staphylococcal infections. Since starting the use of local mupirocin more than 10 years ago, we have had no patients experiencing infections due to the buttonhole technique. The buttonhole technique should not be used for grafts. A technique called rope-ladder is usually employed, where a different hole is used with every dialysis.

Although observational studies suggested equal or decreased vascular access complications during daily dialysis, the Frequent Hemodialysis Network trial (FHN), a randomized controlled trial (RCT) sponsored by NIH and CMS, reported increased access intervention in the daily HD patient group as compared to controls. This does not necessarily imply decreased access survival.

Safety is of paramount importance when using AV access, especially during the night on NHD. Disconnection of the “venous return” needle may not be sensed by the dialysis machine, leading to exsanguination. To ensure the safety of the procedure, adequate taping of the needle and anchoring of the blood tubing are essential. An additional alarm mechanism is highly recommended (moisture sensor) to awaken the patient in the case of blood extravasation from the access site. Most centers use an inexpensive non-disposable “enuresis alarm” for this purpose.

A blood leak detector system has become commercially available for this purpose, using disposable sensors attached to a non-disposable alarm device (RedSense Medical AB). Furthermore, a dialysis machine has become commercially available (Fresenius Medical Care, Waltham, MA), which incorporates a wireless sensor for blood leak, leading to blood pump arrest.

The use of single needle dialysis technique is encouraged for patients on NHD. It can provide adequate blood flow while it decreases the likelihood of complications related to accidental needle dislodgement and bleeding during hemodialysis.

Catheters

Dialysis catheters can be used successfully for intensive hemodialysis. As low blood flow is adequate for NHD, even a poorly functioning catheter is often adequate. The use of pre-perforated dialysis catheter caps is of significant importance during NHD. There are at least three products available: The InterLink® (Becton-Dickinson, Utah, USA), the TEGO® (ICU Medical, CA, USA) and the Swan-Lock® (Codan, Lensahn, Germany) systems have been used.

Dialysis is performed without removing the catheter caps. Resistance to flow provided by these attached caps (dissimilar among the different brands) does not prevent attaining adequate dialysis, as low blood flow is adequate for long hemodialysis. These attachments prevent air embolism in the case of accidental disconnection of the tubing from the dialysis machine. Bleeding is also prevented if the “arterial” but not the “venous” port is disconnected. Therefore, meticulous attention should be given to the proper taping of the tubing to the catheter to prevent disconnection, especially of the venous port.

Complications

Thrombosis:

The local instillation of 2 mg of alteplase (tissue-type plasminogen activator or tPA) has been very successful in restoring adequate blood flow in obstructed catheters due to thrombin. The instillation of lyophilized tPA has been practiced successfully by patients at home.

The approach to the prevention of catheter thrombosis is variable in different centers. Some centers use a small dose of warfarin to prevent catheter thrombosis, while others use prophylactic tPA. The initial warfarin dose is usually 2 mg/day. For patients with recurrent thrombosis, the dose is increased by 1 mg until patency is maintained, providing the INR remains below the levels that the physician considers safe -usually below 2. However, a prospective randomized controlled study did not find the use of low dose warfarin useful in preventing catheter dysfunction. The risk of bleeding, vascular calcification related to the use of warfarin, and the negative trial cited should be considered for individual patients.

Infection:

The incidence of catheter-induced bacteremias with nocturnal hemodialysis is relatively low (approximately 0.35 to 1.5/1000 days). A recent review by Hayes of 98 patients on NHD using catheters during a study period of 6.5 years, found that 64.4% of patients developed bacteremia with a rate of 1.12 episodes per 1000 days. The most common organism was staphylococcus epidermidis (51.4%). At our center, patients with fever and chills during the initiation of dialysis but without severe symptoms, such as hypotension, usually draw blood cultures at home and begin empiric treatment with intravenous antibiotics (usually 1 gm of Vancomycin and 1.5 mg/kg of tobramycin if they have previously tolerated such agents). The treatment is subsequently modified based upon culture results and continues for two weeks. If appropriate, based upon culture results, continuation of treatment with cefazolin rather than vancomycin is encouraged. If the infection recurs (in our experience usually one month later), the catheter is replaced over a guidewire after restarting or continuing antibiotic coverage. A different site should be chosen if an exit site is present.

Many bacteremias resolve successfully with antibiotics. We immediately remove the catheter with severe sepsis especially if associated with hemodynamic instability or suspicion of metastatic infection. The infection rate and survival of the catheters on nocturnal hemodialysis may be better than conventional hemodialysis, but data from randomized controlled studies are absent.

Practical aspects

One or two pairs of buttonholes are usually created on each fistula. In the latter case the pair of buttonholes is rotated after each hemodialysis session. If a buttonhole is not functional, another buttonhole is created. This can be done by the patient, or by the nurses in the dialysis center, which requires one or more patient visits.

The “BioHole device” (Nipro Co., Osaka, Japan) has been described for the creation of the buttonholes. It involves the insertion of a peg into the fistula hole after the needle is removed, thereby keeping the buttonhole open until the next hemodialysis. This routine is continued for the first 14 days of buttonhole creation, followed by the use of the routine “blunt” needles.

Some patients need sharp needles to cannulate the buttonhole as the healing process is fast and the buttonhole is not functional even 1 day later.

Anticoagulation

Routine anticoagulation may be used both for daily or nocturnal hemodialysis. Either unfractionated (1000 units per hour) or low molecular weight heparin (LMWH) can be used similar to that used in conventional hemodialysis. LMWH has been used successfully with Nocturnal Hemodialysis. In the case of heparin-induced thrombocytopenia (HIT), argatroban and danaparoid have been successful alternatives, however the cost is prohibitive.

Diet

Intensive hemodialysis provides better uremic control compared to CHD. Fewer dietetic restrictions are needed in the case of daily hemodialysis. There are no dietetic restrictions while on daily NHD for most of the patients.

Need for supplements

NHD leads to removal of 10-15 gm of aminoacids per session. Therefore, the patients are advised to follow a high protein diet; no ill-effects have been described related to the amino acid losses. The serum amino acid concentration seems to improve on NHD.

Water soluble vitamins are recommended for the patients on intensive hemodialysis. Although there is no adequate data available, most centers suggest two multivitamin tablets for patients while on NHD. Some clinicians suggest administration of vitamin C on top of the amount available in the multivitamin preparation. We have prescribed vitamin C 500 mg daily in selected patients with low serum vitamin C levels. The vitamin C assay is sensitive to suboptimal handling of the sample and this leads often to falsely low values.

Patient selection

Intensive hemodialysis can be offered to alleviate patient symptomatology, improve outcomes and to promote patient independence and rehabilitation.

Symptomatic patients

Symptoms related to uremia, dialysis or comorbidities can improve on intensive hemodialysis. Patients with left ventricular hypertrophy (LVH), refractory heart failure, uncontrollable hypertension, ascites, malnutrition, amyloidosis or neuropathy have been treated successfully with short daily or nocturnal hemodialysis. Facility-based SDHD and NHD three nights a week have been used to treat patients who are failing on conventional HD. Hemodynamic instability and low BP during dialysis is not a contraindication but rather an indication for NHD. Daily NHD is often the only hemodialysis schedule that can be tolerated by very hemodynamically compromised patients.

Healthy patients wishing to improve their quality of life

Patients choose intensive home hemodialysis to improve quality of life and enhance physical and vocational rehabilitation. Patients who cannot be transplanted, want to work, live away from dialysis facilities or have difficulty adjusting to the disciplined life of conventional facility-based HD, should be encouraged to choose home hemodialysis. Other good candidates include patients who need to transfer from CAPD to HD.

Many dialysis centers have adopted a “Home First” policy, where patients are educated and encouraged to choose a home-based dialysis method, either PD or home hemodialysis. If patients with chronic kidney disease (CKD) are educated early enough about the benefits of home dialysis, recruitment to home dialysis is more successful than for a prevalent patient on facility-based dialysis. Resources to help patients and their families to choose the appropriate home dialysis modality have been created (e.g., MATCH-D; available at www.homedialysis.org/match-d).

Detailed aspects of patient selection

Patients with severe cardiomyopathy associated with hypotension have been dialysed using daily NHD at home despite predialysis systolic blood pressure (BP) lower than 100 mm Hg. Most of these patients improved over time with progressive decrease of their extracellular fluid volume (ECFV). This improvement, in most cases, is striking. A common clinical dilemma is the following: in the presence of hypotension, should preference be given to fluid removal or to maintenance of treatment with RAS inhibitors and beta-blockers? Although sufficient data is not available, we have found that fluid removal and temporary discontinuation of the medications is most effective.

Need for helpers

Currently no firm policy exists regarding the need for a partner for frequent or long home hemodialysis. There have been many patients who dialyzed on NHD without the presence of a partner. However, hemodynamically unstable patients, especially those on shorter dialysis schedules, may require the presence of a partner. Several patients who cannot be trained for home hemodialysis have been dialysed with the help of unpaid (usually family members) or paid helpers. The expense for paid helpers is usually borne by the patients or their family. Helpers without previous dialysis knowledge or experience have been successfully trained and represent the majority of the paid helpers in our institution.

Contraindications to intensive home hemodialysis

Patients with poor insight, untreated psychosis, or uncontrolled seizure disorder or brittle DM with frequent hypoglycemias may not be able to do home HD. Furthermore, patients who are unwilling and unable to learn how to do dialysis or handle emergencies are obviously poor candidates. The presence of a helper who has been trained to perform dialysis is a good alternative. Inability of the patient or the helper to communicate with the nurse, the physician on-call or the Emergency Medical Services due to language barriers would also rule out home hemodialysis. Contraindication to systemic anticoagulation is the only absolute contraindication and applies solely to NHD.

Barriers to home hemodialysis

There are several patient-related barriers to home HD. Even if patients have the willingness and ability to learn. These included difficulty with self-cannulation, the fear of a catastrophic event, and the burden on the family.

Barriers related to financial reasons are significant. These include inadequate reimbursement for more frequent dialysis as well as expenses incurred by the patients dialysed at home. Patient borne expenses include electricity, water, sewage, and waste disposal costs. The amount incurred by patients in our center is as high as $1500-2000 per year.

In a few instances, the patient is unable to master the fistula cannulation technique. It is our opinion that using a dialysis catheter and persevering with an intensive hemodialysis at home rather than switching to facility dialysis is the preferred approach. There is insufficient evidence supporting this approach.

Training for home hemodialysis

Ideally, incident patients who are candidates for home HD should be kept separate from the prevalent hemodialysis patients until the training starts, as they often settle on CHD and do not choose home HD. The use of a “transition unit” where patients start on dialysis away from the rest of the established dialysis patients has been practiced in some centers with considerable success. The patient adaptation can be smoother and the choice of home dialysis more likely. The patient or helper undergoes a graded learning program co-ordinated by a dedicated allied staff including training nurses, technologists, pharmacists, dieticians, social workers and physicians. The patients trained for nightly dialysis should dialyse once or twice at night in the center after the completion of their training. This approach has not been universally practiced.

Remote monitoring of the dialysis machine, if available, for 3 months, increases the comfort of patients. A home visit by the training nurse on the first day of discharge home is useful. Some programs conduct a formal assessment of the patient/helper at the end of the training, which leads to “certification of competence.” A yearly recertification is advised by some centers.

The home dialysis patients are encouraged to have adequate home insurance to cover costs related to dialysis-related damage, usually water leaks. Some centers would not allow home HD to proceed without home insurance coverage.

Home dialysis patients are requested to do laboratory tests monthly and visit the dialysis clinic every 1-4 months depending on their clinical stability. Post-HD laboratory tests are recommended on the following: potassium, calcium, and phosphorus. The rest of the laboratory tests are similar to those for CHD patients.

The patient/helper training is done while the patient is dialysed either thrice weekly or daily. In the case of daily dialysis, the training time is shorter. Home dialysis training lasts 5-8 weeks. Fistula training is a significant component and adds to the length of the training.

Other aspects of training

Above and beyond the obvious aspects of training for home hemodialysis, most centers train patients how to collect blood and centrifuge the samples to be delivered to the laboratory, self-administer antibiotics and intravenous iron, and collect water samples for testing.

Among the home supplies the following should be considered: Antibiotics enough for treatment for a small number of days (Vancomycin and Tobramycin or ceftazidime in the presence of residual kidney function, are supplied in our center), blood culture bottles, blood collection tubes, intravenous iron preparation (Venofer® is used in our center), Kayexalate® and lyophilized tissue plasminogen activator (Alteplase-TPA) for some patients with dialysis catheters, especially those living away from the center.

Parathyroid hormone testing is challenging as the PTH levels decay quickly in the tube. This is particularly problematic in the case of home NHD where the sample is harvested in the evening. It has been observed that if the blood is collected in tubes containing EDTA and stored in the refrigerator overnight without centrifugation the accuracy of the assay is acceptable. Alternatively, PTH is measured during clinic visits.

The timing of the monthly laboratory tests varies among the centers. As the main risk on daily NHD is low rather than high levels of solutes, we recommend that the laboratory tests are done after a few consecutive nights on dialysis.

What happens to patients with kidney disease undergoing daily and nocturnal dialysis?

Clinical outcomes of intensive hemodialysis

Blood pressure

All intensive HD schedules have been reported to improve BP control. Long intermittent HD for 8 hours 3 times a week, as practiced in Tassin, France for more than 40 years, improved BP control with few or no medications. Decrease in the post dialysis weight and low salt diet were the cornerstones of this approach. BP control with fewer medications or no medications on SDHD and NHD was described in several studies. Improvement in BP control was confirmed in two RCTs: The Alberta trial on NHD and the FHN trials on both SDHD and NHD.

How to control blood pressure

Improvement of BP on the intensive HD regimens is achieved initially through the progressive decrease in post dialysis weight. In some patients, there is a delay of a few weeks in achieving excellent BP control (the “lag phenomenon”). This is consistent with evidence suggesting that vasodilatation is one of the late mechanisms responsible for hypertension control on NHD.

Our approach to BP control on daily HD is similar to the approach followed by the Tassin group. The post dialysis weight of the patients is decreased progressively until BP is well controlled. Antihypertensives are discontinued progressively. As the decrease in the post dialysis weight often leads to reflex tachycardia, beta blockers are discontinued last. Aggressive decrease in the dry weight may accelerate residual renal function loss as shown in the FHN trial.

Regression of LVH

Left ventricular hypertrophy (LVH) mainly in the form of dilated cardiomyopathy affects 70-80% of the dialysis population and has been used as a surrogate for patient survival in both the renal and general population. Regression of LVH has been described in several observational studies on SDHD and NHD but not as conclusively on intermittent long hemodialysis.

Regression of LVH has been used as a surrogate outcome of patient survival in two RCTs. It was detected after 6 months on NHD in the Alberta study when compared to conventional HD. In the FHN trial, frequent hemodialysis, as compared with conventional hemodialysis, was associated with favorable results with respect to the composite outcomes of death or change in left ventricular mass and death or change in a physical health composite score. The regression of LVH was proportionate to its severity. The NHD arm of the FHN trial did not confirm a statistically significant LVH regression in this population despite a positive trend. This arm of the trial was unable to recruit the initially planned number of patients and also accepted a high number of incident patients with higher residual kidney function. These two factors have been offered as a possible explanation for the lack of significant results.

Improvement in cardiac function and endothelial function

Impressive improvement of the ejection fraction of patients with cardiomyopathy on NHD has been described in an observational study. There was improved endothelial function and decrease in sympathetic tone.

Patients on dialysis are characterized by impaired endothelial function, autonomic system imbalance characterized by increased sympathetic tone, leading to vasoconstriction and decreased heart rate variability. The number and function of the progenitor endothelial cells are decreased. These parameters have been associated with increased cardiovascular morbidity and mortality in the renal as well as in the general population.

Improvement of endothelial function was seen on NHD. Norepinephrine levels decreased and the post-ischemia vasodilatory response of the radial artery improved. Peripheral vascular resistance measured by echocardiographic techniques also decreased. Furthermore, the number and function of endothelial progenitor cells improved after conversion from CHD to NHD. Endothelial stem cells harvested from NHD patients improved the ischemic rat hind leg ischemia as compared to poor response to cells derived from patients on CHD.

The FHN trial confirmed that the vagal tone increases and heart rate variability increases on frequent HD.

Some of the improvements in endothelial function correlated temporally with improvement of serum phosphate levels, supporting the hypothesis of cardiovascular phosphate toxicity. Other benefits observed on NHD included improvement in baroreflex function, exercise duration and capacity and improvement in muscle cell biology.

Quality of life

There are several observational studies reporting improved quality of life on intensive hemodialysis including long intermittent hemodialysis, SDHD, and daily NHD. There are two RCTs published that examined the treatment effects on QOL. The Alberta study randomized 52 patients to either CHD or home NHD for 6 months. The improvement in QOL on NHD did not achieve statistical significance (p=0.06). Several kidney-specific quality of life domains showed significant improvement on NHD (p=0.01).

The FHN trial randomized 250 patients to CHD or facility-based SDHD. There was a significant improvement of the co-primary composite outcome of death or change in the physical-health composite score of the RAND 36-item health survey. Such an improvement was not seen in the nocturnal arm of the trial. However, there was a trend for improvement in the health survey in both study and control groups of patients. This may be related to the benefit of initiation of dialysis, as most of the patients were incident patients, or to the change in the dialysis location from in-center to home.

Residual kidney function

In the FHN trial the patient residual renal function decreased more rapidly in the nocturnal arm of the trial compared to controls. It should be noted that the patients recruited for the NHD arm of the trial had higher residual kidney function than in the daily HD arm.

Mineral metabolism

Phosphate control

Hyperphosphatemia is a significant problem in CKD and has been linked to both CKD metabolic bone disease as well to vascular calcification, increased morbidity, and mortality. As described above, SDHD is associated with improved phosphate control if the duration of dialysis per week exceeds the time on conventional HD. Three-hour daily dialysis has been recommended. This was confirmed on the FHN trial, a NIH sponsored RCT. Phosphate control is excellent on daily NHD resulting in the discontinuation of phosphate binders in all patients. Moreover, there is the need to add phosphate into the dialysate in about 75% of the patients. Three times a week or every other night NHD is also effective with 80% of the patients not needing phosphate binders. Obviously, phosphate intake is less restricted on SDHD and unrestricted on NHD. High phosphate intake is encouraged on daily NHD.

Calcium balance

Calcium balance can be negative on long hemodialysis. Calcium loss through ultrafiltration of the ionic calcium component is proportionate to ultrafiltration volume, therefore proportionate to the salt and water intake. Most patients on intensive dialysis use lower dose or no calcium-based phosphate binders and also have low GI calcium absorption due to low serum vitamin D levels. All the above may lead to negative calcium balance. Therefore, dialysate calcium needs to be higher than on conventional hemodialysis to maintain calcium balance.

In most patients on daily NHD dialysate calcium needs to be greater than 3 mEq/L (1.5 mmol/L). Dialysate calcium should be adjusted aiming to maintain serum post-dialysis calcium higher that the pre-dialysis value (to accommodate the increase in the total calcium concentration due simply to the hemoconcentration and effect on the albumin-bound calcium).

Parathyroid hormone (PTH)

PTH levels are a function of dialysate calcium levels. Usually adjusting the dialysate calcium levels will allow the PTH to be within the desirable range. There is no information as to what the ideal PTH should be on intensive hemodialysis; therefore, KDOQI guidelines should be followed. Changes in alkaline phosphatase are useful in selecting the appropriate PTH levels. In the absence of hyperphosphatemia, post-dialysis transient hypercalcemia necessary for control of elevated PTH levels may be safe, especially in patients without evidence of arterial calcifications. In the presence of vascular calcifications, hypercalcemia should probably be avoided and the use of calcimimetics or parathyroidectomy should be considered.

Phosphate additive

The need for phosphate additive into the dialysate should be based on the pre- and post-dialysis serum phosphate. Ideally laboratory tests should be done on a steady state while on daily HD (after 3 consecutive nightly sessions) but after the longer interdialytic interval on intermittent dialysis. The timing of the testing should be taken into account in the dialysate composition prescription.

Phosphate should not be added in most patients during dialysis after a long interdialytic interval. The amount of Fleet enema can be safely increased above the above recommended levels (Table 1) in cases of bone formation (post parathyroidectomy, while on calcimimetics especially in the presence of high bone-derived alkaline phosphatase or during pregnancy). Phosphate does not precipitate with calcium in the ‘acid’ concentrate due to the presence of acid pH.

Fleet enema doses more than 200 ml per jug have been occasionally employed. In the case of long intermittent hemodialysis it is often difficult to normalize both pre- and post-dialysis P because correction of post-dialysis hypophosphatemia can lead to high pre-dialysis P values. This can be resolved by accepting serum pre-dialysis phosphate up to 4.5 mg/dL (1.5 mmol/L) while ignoring mild post dialysis hypophosphatemia.

Increasing levels, or lack of correction of high PTH levels, or are increasing or not decreasing elevated alkaline phosphatase levels may be suggestive of inadequate dialysate calcium levels.

Flexible calcium dialysate composition can be achieved through calcium chloride powdered additives; 15 mL of powder increases dialysate calcium by 1 mEq/L (0.5 mmol/L).

This allows the continuous adjustment of the dialysate composition for home hemodialysis without the need to deliver a new dialysate supply to the home of the patient. The patients in our center use a volumetric cylinder (empty syringe) to measure the volume of the calcium chloride powder. The mistaken addition of calcium chloride into the bicarbonate bath will cause precipitation of calcium carbonate and lead to hypocalcemia, high PTH and increased levels of alkaline phosphatase. Dialysate calcium concentrations more than 4 mEq/L or 2 mmol/L have been occasionally needed in cases of ‘hungry bone’ syndrome.

Sleep

Sleep disorders are very common and under-recognized in patients at all stages of CKD. The most common disorders include sleep apnea, insomnia, excessive sleepiness, restless leg syndrome and periodic limb movement disorder. The prevalence of sleep apnea in the dialysis population is higher than 50%. Conversion of patients from conventional hemodialysis to daily nocturnal hemodialysis was associated with a significant decrease in the apnea-hypopnea index and an improvement in the hypoxemia during sleep. Similar effects were observed on nightly peritoneal dialysis (APD). The effect of SDHD on sleep disorders is unknown. In a recent study home hemodialysis was associated with improvement in the severity of the restless leg syndrome.

Insomnia is not uncommon during the early period on nocturnal hemodialysis. The judicious use of short acting sedatives has been used successfully in some patients. This can be considered in the presence of a helper at home, who would be able to react to dialysis machine alarms.

Sleep-mechanistic analysis

Sleep apnea is caused by the destabilization of central ventilatory control and upper airway occlusion during sleep as well as by enhanced ventilatory sensitivity to hypercapnia. These abnormalities seem to improve with the conversion to nocturnal hemodialysis. Upper airway occlusion can be caused by fluid overload and interstitial edema in the upper airway. Conversion of ESRD patients from CHD to NHD was associated with an increase in pharyngeal cross-sectional area, possibly due to improved fluid removal. Similarly, conversion from CAPD to nocturnal peritoneal dialysis is associated with reduced pharyngeal narrowing as measured by magnetic resonance imaging.

Nutrition

In observational studies, conversion of patients from CHD to SDHD was associated with increased body weight, nPNA serum albumin levels, and arm muscle area. Anecdotal experience with patients on both regimens suggests increase in appetite and at times very significant weight gain. Hemodialysis is associated with amino acid losses in the dialysate, which would be higher on more frequent and long hemodialysis. No negative effects have been reported but a relatively higher protein diet is recommended in these patients. Long term nutritional stability was reported on long hemodialysis while decrease in body weight with time has been associated with the use of CHD. The FHN trial did not show significant change in body composition on frequent HD. Nonetheless, more frequent dialysis regiments liberate dietary intake of salt, water, potassium, phosphate and protein. Although potassium intake is not restricted, the amount needs to be relatively consistent. The dose of multivitamin preparation is increased to two tablets daily. However, there has not been any evidence of vitamin deficiency.

Erythropoiesis-stimulating agents and iron administration

The reported effects of SDHD and NHD on anemia and the dose of erythropoiesis stimulating agents (ESAs) have been variable. Several observational studies reported increasing hemoglobin levels as well decreasing need for ESA. The three RCTs published to-date failed to demonstrate the effect of daily or nocturnal hemodialysis on anemia control. The trial population characteristics and the short time on the studies may be relevant to the results.

In our center, a decrease in the dose of ESA by about 30% was observed less than 1 year after the initiation of NHD. About 25% of the patients on NHD in our center are off ESA. In vitro, growth of erythroid (BFU-E) and granulocytic (CFU-GM) colonies was superior when cultured with NHD plasma compared with conventional HD plasma, supporting the beneficial effect of intensive dialysis. Most centers train patients on home dialysis to self-administer intravenous iron usually in the form of iron dextrose (Venofer®).

The first dose of Venofer® is administered in a facility capable of treating anaphylaxis. We did not encounter significant adverse reactions to Venofer® administration at home over the last 15 years.

The dose of intravenous iron is similar to CHD. An increased dose is often needed during pregnancy. Most patients are given Venofer 100 mg intravenously twice a month. Several patients off ESA have hemoglobin levels above 13 gm/dL or 130/L. The need to replenish low iron stores in these patients with hemoglobin above the DOQI guideline levels off ESA is unclear.

Effect on reproductive biology

Fertility is decreased in ESRD and pregnancy is often associated with poor outcomes. Intensive HD has been used over the last few years to improve fertility rates and pregnancy outcomes. Improved hormonal function has been described on daily hemodialysis. Several successful pregnancies have been reported on daily NHD; therefore, intensive hemodialysis has been advocated prior to and during pregnancy. Use of daily NHD is the preferred dialysis modality during pregnancy, based only on observational studies and one patient cohort comparison. Alternatively, other forms of intensive hemodialysis should be used. Patients dialyzed for more than 36 hours per week had the highest rate of live births (85%).

The need for increased amounts of calcium and phosphorus in the dialysate has been observed and is likely related to the growth of the fetal skeleton. One should be guided by serum pre- and post-dialysis calcium and phosphorus as well as PTH. If the deficits are not corrected, elevated levels of PTH and alkaline phosphatase will be evident.

The need for ESA and iron supplements during pregnancy is higher. The anemia may be partially related to increased intravascular volume expansion associated with pregnancy which is accommodated by the dialysis prescription as dictated by clinical parameters.

Some patients develop hypertension during pregnancy. Pharmacological treatment is similar to that of pregnant women with normal renal function. Labetalol, alpha methyldopa, and other appropriate medications have been used.

The safety of aggressive ultrafiltration for blood pressure control is unclear. Judicious use of ultrafiltration in the presence of symptoms of breathlessness or significant edema is reasonable. The effect or ultrafiltration on placental flow is not known. Most clinicians tend to avoid using ultrafiltration as the only measure to control hypertension.

If hypertension continues to be a problem despite the use of antihypertensives, especially in the presence of extracellular volume expansion, intensification of ultrafiltration may be acceptable. In the postpartum period, the more liberal use of ultrafiltration is useful in controlling blood pressure, allowing for the decrease in the dose or discontinuation of the medications. It has been speculated that aggressive ultrafiltration in the post-partum period may affect milk production negatively, although this has not been well documented.

The level of dialysate calcium to sufficiently prevent a significant rise in PTH is usually higher than 3.5 mEq/L or 1.75 mmol/L. Often, the amount of intradialytic phosphorus needs to be increased to as high as 200 ml of Fleet® enema per dialysate jug. The amount is adjusted to ensure that both the pre- and post- dialysis serum phosphate are maintained normal.

The ideal level of hemoglobin is unknown. Intravenous iron is administered as needed to maintain a transferrin saturation of 0.25 or higher. This usually requires an increased dose of intravenous iron.

Pregnancy is normally associated with respiratory alkalosis, which in turn is associated with lower serum bicarbonate as part of the kidney compensation. Some clinicians suggest decreasing the dialysate concentration of bicarbonate during pregnancy to achieve a similar effect. The level can be adjusted to maintain a predialysis serum bicarbonate at 20-22 mEq/L. No specific data exist to support the notion. Several pregnancies were successful without this adjustment.

Other effects of Frequent HD as described in the FHN trial

Frequent hemodialysis did not improve executive function or global cognition in the FHN trial.

Frequent in-center hemodialysis compared with conventional in-center hemodialysis improved self-reported physical health and functioning but had no significant effect on objective physical performance. There were no significant effects of frequent nocturnal hemodialysis on the same physical metrics.

Cost

The direct cost of more frequent dialysis is higher than CHD. The financial impact of intensive HD depends on the stakeholder’s perspective. Some significant stakeholders include the dialysis providers, payers for dialysis, hospitalization, medication costs), patients, and society at large. Home dialysis includes higher capital costs but lower labor cost. The most expensive intensive hemodialysis modality is in-center SDHD, where the cost of both consumables and labor increase.

Ultimately, the reimbursement structure of the specific jurisdiction dictates the financial incentives or disincentives for the adoption of home daily intensive dialysis methods. Several observational studies in Canada, Australia, and the US suggest that the overall cost of patient care on home SDHD or daily NHD is lower than the cost of CHD, if the calculation includes hospitalization and medication costs. A cost utility study also found that the cost of QALY (Quality Adjusted Life Years) was lower on NHD as compared to CHD.

The reimbursement structure and rates in the US have been unfavorable for the use of daily home hemodialysis. This may have changed somewhat recently with the adoption of the “bundling” of the reimbursed items to include intravenous medications such as ESAs and vitamin D analogues, the use of which may be lower on intensive HD. There is no RCT comparing the cost of intensive to conventional dialysis. In-center intermittent NHD may be financially attractive depending on the intensity of labor utilization.

Current utilization

There are no accurate data on the utilization of intensive HD. Data from the International Quotidian Dialysis Registry suggest that the distribution of the intensive HD regimens differs in the different jurisdictions. The most frequent modality in Australia/New Zealand is every other day NHD at home, while in Canada daily NHD and in the US in-center NHD is the most common modality. The utilization of SDHD is increasing in the US, using mainly the NxStage dialysis system.

Patient survival

There are no RCT-derived data regarding patient survival on intensive HD. The complexity and the cost prohibit the likelihood that the study will ever be done.

Observational studies have described patient survival on the different schedules. Selection biases include younger age for the home HD patients (by about 10 years) and lower comorbidities.

Most of these studies, irrespective of intensive HD modality, have reported patient survival rates of about 80% over 5 years. A retrospective matched cohort study found similar survival rates with nocturnal hemodialysis in Canada and deceased donor transplantation (US data). In another study, deceased donor transplantation patient survival rates were higher than on nocturnal hemodialysis when only Canadian data was used.

Although patient mortality rates were not the primary end-points of the FHN trial, the 12-month frequent in-center hemodialysis significantly reduced the long-term mortality over a median of 3.6 years. Conversely patients assigned to nocturnal hemodialysis in the FHN trial had a higher mortality rate especially during the year after the completion of the trial; the low number of patients recruited and the surprisingly low mortality rate in the control group (0.03 deaths/patient year) make these results difficult to interpret.

Outcomes of Daily hemodialysis versus other modalities

In several studies based on institutional databases from Australia/New Zealand and North America, frequent and home hemodialysis had better patient survival rates, fewer hospital admissions and fewer conversions to in-center hemodialysis than peritoneal dialysis.