Crit-Line In-Line Monitor

Number: 0373

Table Of Contents

Policy
Applicable CPT / HCPCS / ICD-10 Codes
Background
References

---

Policy

Scope of Policy

This Clinical Policy Bulletin addresses Crit-Line in-line monitor.

Experimental and Investigational

Aetna considers the Crit-Line In-Line Monitor (In-Line Diagnostics, Kaysville, UT) experimental and investigational for monitoring hematocrit and oxygen saturation of members because the effectiveness of the Crit-Line In-Line Monitor in avoiding intra-dialytic morbid events, in preventing ischemia due to intra-dialytic hypoxia, and in determining access re-circulations has not been proven.

---

Table:

CPT Codes / HCPCS Codes / ICD-10 Codes

Code	Code Description
Information in the [brackets] below has been added for clarification purposes.   Codes requiring a 7th character are represented by "+":
Crit-Line In-Line monitor:
No specific code
Other CPT codes related to this CPB:
90935	Hemodialysis procedure with single evaluation by a physician or other qualified health care professional
90937	Hemodialysis procedure requiring repeated evaluation(s) with or without substantial revision of dialysis prescription
Other HCPCS codes related to this CPB:
S9335	Home therapy, hemodialysis; administrative services, professional pharmacy services, care coordination, and all necessary supplies and equipment (drugs and nursing services coded separately), per diem
ICD-10 codes not covered for indications listed in the CPB (not all-inclusive):
N17.0 - N17.9	Acute kidney failure
N18.1 - N18.9	Chronic kidney disease (CKD)
N19	Unspecified kidney failure
N26.1, N26.9	Unspecified contracted kidney
Z49.31	Encounter for adequacy testing for hemodialysis
Z49.32	Encounter for adequacy testing for peritoneal dialysis
Z99.2	Dependence on renal dialysis

---

Background

The Crit-Line In-Line Monitor has been used in hemodialysis patients and other fluid-overloaded patients to measure hematocrit levels and oxygen saturation.  Crit-Line In-Line Monitor optically measures hematocrit and oxygen saturation as blood passes through the dialysis tubing while a patient is undergoing dialysis treatment.  The Crit-Line graphically displays and records the hematocrit, the percent blood volume change (which is derived from the hematocrit), and the oxygen saturation.  The hematocrit and oxygen saturation are measured by an optical sensor attached to a sterile disposable blood chamber that is placed in-line between the arterial blood tubing set and the dialyzer.  The Crit-Line determines the hematocrit based on both the absorption properties of hemoglobin and the scattering properties of red blood cells passing through the blood chamber.

The efficacy of the Crit-Line In-Line Monitor in avoiding intra-dialytic morbid events, in preventing ischemia due to intra-dialytic hypoxia, and in determining access re-circulations has not been proven.

The Crit-Line In-Line Monitor (In-Line Diagnostics Corporation, Riverdale, UT) is used to measure a patient's hematocrit level and oxygen saturation during hemodialysis.  The Crit-Line In-Line Monitor has been used in hemodialysis patients and other fluid-overloaded patients.  For patients without functioning kidneys, periodic hemodialysis is performed to remove toxins and to remove excess fluid.  Between hemodialysis treatments, patients may experience hypertension as a result of fluid overload, whereas during hemodialysis treatment, patients may experience hypotension due to excessive plasma volume depletion.  There is evidence that volume overload (Mees et al, 1995) and hypertension (Mailloux et al, 1994) may affect the incidence of cardiovascular disease in hemodialysis patients.

The Crit-Line In-Line Monitor optically measures hematocrit and oxygen saturation as blood passes through the dialysis tubing while a patient is undergoing dialysis treatment.  The Crit-Line graphically displays and records the hematocrit, the percent blood volume change (which is derived from the hematocrit), and the oxygen saturation.  The hematocrit and oxygen saturation are measured by an optical sensor attached to a sterile disposable blood chamber that is placed in-line between the arterial blood tubing set and the dialyzer (Steuer et al, 1993).  The Crit-Line determines the hematocrit based on both the absorption properties of hemoglobin and the scattering properties of red blood cells passing through the blood chamber.

The device was cleared for marketing by the Food and Drug Administration (FDA) in January 1995, based on a 510(k) application showing substantial equivalence of the Crit-Line to other devices on the market (Callahan, 1995).  Thus, the manufacturer did not need to provide the FDA with data on the efficacy of the Crit-Line device. 

The manufacturer claims that the Crit-Line’s in-line method of measuring hematocrit is an advance over current methods of determining hematocrit, including micro-centrifugation, electrical conductivity measurements, and photometry.  According to the manufacturer of the Crit-Line, use of the micro-centrifuge does not allow for monitoring of real-time changes in blood volume (Steuer et al, 1993).  Taking blood samples every 10 to 15 minutes is labor-intensive, and the measurements are not precise.

Measurements of blood electroconductivity allow for real-time hematocrit measurements, but the results are affected by abnormal electrolyte, anti-coagulant, and protein concentrations (Steuer et al, 1993).  If correction is not made for these abnormal concentrations, hematocrit readings will be inaccurate.

Optical measurements photometry allow for real-time hematocrit measurements, but readings are affected by ambient light, tubing artifact, changes in blood-flow rate, in-line pressure, and oxygen saturation (Steuer et al, 1993).  These optical measurements measure “relative hematocrit” and have to be recalibrated with each use.  By contrast, the Crit-Line measures “absolute” hematocrit and does not need recalibration.

One study by Steuer et al (1993) assessed the accuracy and precision of Crit-Line hematocrit measurements in 4 volunteers, comparing Crit-Line measurements with hematocrit values obtained by microcentrifugation.  A total of 15 dialysis sessions were monitored during the study, resulting in 63 sets of paired data points.  Linear regression showed a correlation of 0.996 between Crit-Line hematocrit measurements and hematocrit measurements by microcentrifuge, with 95 % confidence interval (CI) (2 standard deviations [SD]) of +2.32 to -2.32 hematocrit percent.  The precision of the Crit-Line measurements was evaluated by comparing repeated measurements.  The 95 % CI was +0.56 to -0.56 hematocrit percent.

A subsequent study compared Crit-Line hematocrit measurements to hematocrits measured by micro-centrifuge (Steuer et al, 1994).  The correlation coefficient was 0.89 and the 95 % CI was +4.8 to -4.8 hematocrit percent.

The investigators in this study also found some evidence for a hematocrit threshold in some dialysis patients that is related to “intradialytic morbid events” (i.e., adverse events that occurred during dialysis), such as hypotension, muscle cramps, and lightheadedness.  In order to determine whether future intradialytic morbid events can be predicted, Steuer and colleagues (1994) studied the blood pressure, blood volume, and hematocrit changes during dialysis in 8 generally stable dialysis patients and 8 hypotensive-prone dialysis patients.  Each subject was studied on 6 separate dialysis sessions within a 4-week period during which the dose of epoetin remained constant.  Blood pressure and pulse rate were measured by an automated cuff every 15 minutes.  Hematocrit was monitored continuously using Crit-Line measurements.  Hematocrit was also measured by micro-centrifugation before dialysis, hourly during dialysis, and immediately after dialysis.  Change in blood volume was calculated from the observed change in hematocrit.

Intra-dialytic morbid events occurred in about half of the dialysis sessions (Steuer et al, 1994); the high rate of intra-dialytic morbid events was due, in part, to the bias in selection of hypotensive-prone patients for the study.  The investigators determined whether there were significant differences in these parameters between dialysis sessions where intra-dialytic morbid events occurred and dialysis sessions without intra-dialytic morbid events.  They found that ultra-filtration rates were significantly higher during hemodialysis sessions where intra-dialytic morbid events occurred (1,250 ml/hr, SD 300 ml/hr) than during hemodialysis sessions where no intra-dialytic morbid events occurred (950 ml/hr, SD 220 ml/hr).  They also found a difference in the percent change in blood volume between hemodialysis sessions where intra-dialytic morbid events occurred and sessions where no intra-dialytic morbid events occurred; however, this difference did not achieve statistical significance.

In hemodialysis sessions with intra-dialytic morbid events, there was no significant correlation between change in blood volume and ultra-filtration rate (r = 0.04); but in hemodialysis sessions without intra-dialytic morbid events, there was a significant negative correlation between blood volume change and ultra-filtration rate (r = -0.52) (Steuer et al, 1994).  The rate of change of blood volume was significantly higher in hemodialysis sessions where intra-dialytic morbid events occurred (12.2 % per hour, SD 5.5 % per hour) than in sessions where no intra-dialytic morbid event occurred (5.6 % per hour, SD 3.6 % per hour). The investigators concluded, however, that although the rate of change of blood volume would allow one to predict the likelihood of intra-dialytic morbid events, they would not help to predict when they would occur.

Measurement of changes in hematocrit did not reliably predict changes in blood pressure.  The dependence of blood pressure on hematocrit was variable, both from day-to-day and from patient to patient (Steuer et al, 1994).  Inverse correlations between blood pressure and hematocrit were found in only 10 of the 16 patients.

Positing that a patient-specific, critical hematocrit exists that triggers intra-dialytic morbid events, the investigators calculated a "hematocrit threshold" for each patient that had more than 1 intra-dialytic morbid event.  The hematocrit threshold was the average hematocrit level of each patient at the times where intra-dialytic morbid events occurred (Steuer et al, 1994).  They found that, for a given patient, the hematocrit at which an intra-dialytic morbid event occurred seemed fairly constant (the average standard deviation of hematocrit measured at intra-dialytic morbid events was 1.4 %). The investigators posited that a critical hematocrit threshold exists for certain patients, and that intra-dialytic morbid events can be avoided by continuously monitoring the patient's hematocrit with the Crit-Line.

The investigators acknowledged, however, that additional studies would have to demonstrate that altering dialysis based on a patient's hematocrit threshold can indeed reduce intra-dialytic morbidity and enhance the efficiency of fluid removal during dialysis (Steuer et al, 1994).  They would also have to determine whether the hematocrit threshold remains consistent with changing conditions in the patient on dialysis, such as where red cell mass changes due to epoetin therapy or bleeding.

To test this hypothesis, investigators examined the oxygen saturation measurements during the dialysis sessions reported in the previous study (Steuer et al, 1995).  Although the oxygen saturation measurements did not correlate to intra-dialytic morbid events, episodes of hypoxia (defined as an oxygen saturation below 90 %) occurred during at least 1 dialysis session in 9 of the 16 patients.  Two of the 16 patients had significant desaturations during sleep, suggestive of sleep apnea, and 2 others with chronic obstructive pulmonary disease had episodes of hypoxia during dialysis.  The investigators posited that the high incidence of cardiovascular disease in dialysis patients may be due, in part, to silent ischemia which occurs during periods of hypoxia during dialysis.  They suggested that continuous measurement of oxygen saturation during dialysis could help to identify the underlying etiologies of intra-dialytic hypoxia, and diagnose and treat sleep apnea, chronic obstructive pulmonary disease, pulmonary disease, and other complications.  Furthermore, continuous monitoring and appropriate oxygen therapy could be used to prevent intra-dialytic hypoxia and thus reduce its contribution cardiac ischemia and death.

Although this is a plausible hypothesis, the investigators recognized that further study is needed of the contribution of intra-dialytic hypoxia to cardiac disease, and to the effectiveness of continuous monitoring and oxygen therapy to the reduction in cardiac disease incidence in dialysis patients.

A fourth study examined the accuracy of Crit-Line measurement of access recirculation (Germain et al, 1995).  Determination of access recirculation is important in assessing fistula patency and integrity, and for maintaining high dialysis efficiency.  The investigators compared re-circulation measured by Crit-Line with simultaneous measurements of recirculation using a BUN test in 46 dialysis patients.  The investigators found a high correlation of Crit-Line and BUN measurements of recirculation (r = 0.93).  Forty of the patients were found to have less than 8 % re-circulation by Crit-Line measurement, whereas 6 patients were found to have greater than 10 % re-circulation.  All of the patients with re-circulation greater than 10 % were found to have a significant stenosis as determined by arteriography and pressure measurements.  Because the incidence of stenosis in dialysis patients with less than 8 % re-circulation was not assessed, the sensitivity of the Crit-Line in detecting stenosis was not determined.

A subsequent study evaluated the Crit-Line in 5 dialysis patients (Steuer et al, 1996).  Rates of dialysis ultra-filtration were increased by 25 % above prescribed values.  Incidence of side effects with adjustments of continuous adjustment of ultra-filtration rate by Crit-Line to maintain hematocrit above a pre-determined threshhold were compared to incidence of side effects in the same patients in subsequent sessions were there was no continous adjustment of ultra-filtration rate.  There was less symptoms of lightheadedness, cramping, and nausea reported during sessions where Crit-Line was used (26 % versus 57 %, p = 0.038).  The results of this small case series need to be confirmed by larger prospective controlled trials.

Steuer et al (1998) evaluated the efficacy of monitoring changes in blood volume during routine hemodialysis to detect fluid overload.  Intra-dialytic changes in blood volume were monitored by continuously measuring hematocrit using the Crit-Line in all 56 patients in a single dialysis unit over 7 weeks.  After week 1, patients were categorized into 2 separate groups depending on their maximum intra-dialytic decreases in blood volume.  In group 1, 46 of 56 (82 %) had greater than a 5 % decrease in blood volume while in group 2, 10 of 56 (18 %) had less than a 5 % decrease in blood volume.  During weeks 2 to 7, dialytic fluid removal was intentionally increased in group 2 patients by 0.80 +/- 0.62 L (mean +/- SD) or 47 +/- 43 %. This intervention resulted in a larger (p < 0.02) intra-dialytic decrease in body weight (2.7 +/- 0.9 kg versus 2.0 +/- 0.8 kg) and a larger (p < 0.02) intra-dialytic decrease in blood volume (15 +/- 5 % versus 4 +/- 1 %) than experienced during week 1 with a low incidence of symptoms.  The investigators concluded that this study shows that there is a significant percentage of chronic hemodialysis patients who can tolerate additional fluid removal without hypovolemic symptoms even though they are considered to be at dry weight by routine physical examination and that the identification of these patients can be facilitated by intra-dialytic blood volume monitoring.

The Crit-Line is considered investigational because its efficacy in avoiding intra-dialytic morbid events, in preventing ischemia due to intra-dialytic hypoxia, and in determining access recirculation has yet to be demonstrated.

---

References