Despite advances in critical care, adult respiratory distress syndrome contributes to significantly higher rates of morbidity and mortality. Conventional mechanical ventilation can cause high airway pressures and oxygen toxicity. The combination of barotrauma, volutrauma, biotrauma, and oxygen toxicity can exacerbate lung injury from primary illness. Extracorporeal membrane oxygenation (ECMO) is an alternative method providing gaseous exchange, and at the same time, it can protect the severely injured lung from further damage. Extracorporeal membrane oxygenation has been in talk since decades; according to a study published in 1988, the authors concluded that a hollow-fiber oxygenator and a centrifugal pump can provide excellent gas exchange, acceptable hemolysis, and little fluid loss.1 A more recent study focused on the advantages of using the extracorporeal membrane oxygenation for treating severe adult respiratory failure against conventional ventilatory support. This study involved a multicenter randomized controlled trial and was published in 2009. The authors of this study recommended that transferring patients with severe, but potentially reversible, respiratory failure to a center with an ECMO-based management protocol can significantly improve survival without severe disability.2 Patients requiring ECMO support are all critically ill; therefore, fluid balance should be managed as accurately as possible. Some other studies have also focused on fluid loss through neonatal ECMO circuit.3–5
Camacho et al.4 demonstrated that insensible water losses across a silicone membrane correlated with sweep-gas rate but not with membrane size, transmembrane pressure gradient, or pump flow rate. Grist et al.6 described water transfer at 5 and 10 ml/m2 of membrane surface area per hour at 37°C. Heulitt and Pickert7 recommended adding approximately 10% of the calculated maintenance fluid requirements to compensate for the insensible fluid loss through the membrane. In the present study, we used Jostra Quadrox D oxygenator, which is an adult-sized oxygenator with a surface area of 1.8 m2; it uses polymethylpentene closed hollow-fiber technology and has been approved for use by the Federal Drug Administration for 6 hr. Insensible water loss has also been studied in another type of oxygenator, the Hilite 2400LT oxygenator (Medos, Stolberg, Germany), which is also a polymethypentene membrane; there is approximately 72 ml/day of fluid loss per liter per minute gas flow over 24 hr.8
The present study is the first to focus on fluid loss through adult ECMO circuit. The objective of this study is to look at the correlation between fluid loss with different sweep-gas flow rates and fluid temperatures.
In this study, an in-vitro extracorporeal circuit consisting of Jostra Quadrox D oxygenator (Maquet, Germany), a Rotaflow centrifugal pump, heater unit (HU 35) and 3/8 inch tubing to join the components together was used (Figure 1). The manufacturer’s recommended minimum flow rate of the Quadrox D oxygenator is 500 ml/min. It has a static prime volume of 250 ml and surface area of 1.8 m2.
The circuit was primed with normal saline and recirculated at 5 L/min. A 50-ml syringe filled with normal saline was connected to the closed system. It was assumed that all fluid loss occurred via evaporation across the membrane surface area as the system was closed to the atmosphere; fluid loss was reflected by the reduction in the water level within the syringe. A camera was used to record the change in the water level with snapshots taken at an interval of 500 sec over a time period of 100 min. We used short and frequent snapshots at every 500 sec, and we were able to observe the correlation clearly after 100 min of experiment. Water loss was compared among gas flow rates of 3, 5, and 7 L/min and fluid temperatures from 33 to 39°C. The HU 35 allows fluid temperature ranges from 33 to 39°C.
The water loss was compared among the three different sweep-gas flow rates (3, 5, and 7 L/min) and seven different fluid temperatures (33 to 39°C). The average daily water loss for every liter per minute of sweep-gas flow rate was calculated. The correlation between water loss and different sweep-gas flow rates and fluid temperatures were represented as graphs. A regression loss equation was derived from the observation.
The amount of water loss through the ECMO circuit was linearly correlated to the sweep-gas flow rate and fluid temperature (Figures 2 to 5). It was found that for every liter per minute of sweep-gas flow at 37°C, 0.046 ml of water was lost, and for every change of fluid temperature by 1°C, water content loss was changed by 0.0026 ml/min by multiple linear regression (R2 = 0.996). The average daily water loss for every liter per minute of sweep-gas flow at 33, 34, 35, 36, 37, 38, and 39°C were 51.3, 55, 58.8, 62.5, 66.2, 70.0, and 73.7 ml/day, respectively. The amount of daily fluid loss corresponding to different sweep-gas flow rates and fluid temperatures is presented in Table 1.
The following regression loss equation was derived from our observation:
Fluid and electrolytes balance are essential for the well-being of a patient in a critical care setting. Previous studies have focused on insensible water loss through neonatal ECMO circuit since neonates are at an especially high risk due to their low body surface area and blood volume when compared to an ECMO circuit; the ECMO circuits can more than double the blood volume of a neonate. In this study, we hypothesized that insensible water loss through an adult ECMO circuit can be as important as that of a neonatal circuit as all the patients on ECMO are critical in condition and any small change in fluid status and electrolytes balance can have a great influence on patient outcome.
This study used an in vitro ECMO circuit mimicking adult ECMO circuit at a blood flow rate of 5 L/min, and the water reduction in the syringe connected to the closed system reflects the water loss within the ECMO circuit at different sweep-gas flow rates and fluid temperatures. Lawson and Holt3 have demonstrated the Jostra Quadrox D oxygenator has lower insensible water loss per liter of sweep gas than other types of oxygenators described in the recent literature. While Yang et al.9 tested water transmission rates through polyurethane membrane and found that water vapor transmission rate increased with an increase in the dry airflow rate on the membrane and reached its peak at 5 L/min. Yu et al.10 demonstrated that the permanent life support system has partially improved biocompatibility in terms of improved cell preservation, lower trans-membrane pressure drops, less plasma leakage, and thrombus formation. Quadrox D oxygenator and Rotaflow centrifugal pump have been used in adult patients with refractory cardiogenic shock and provided acceptable results in terms of patients surviving on ECMO and discharge.11
In this study, we found that the insensible water loss was linearly correlated with the sweep-gas flow rate and the average daily water losses for every liter per minute of sweep-gas flow at 33, 34, 35, 36, 37, 38, and 39°C were 51.3, 55, 58.8, 62.5, 66.2, 70.0, and 73.7 ml/day, respectively; this result was comparable to previous study results.4
We titrated up the fluid temperature from 37 to 39°C since some patients may develop shivering due to underlying sepsis; hence, titrating up the fluid temperature can eliminate this phenomenon. Also, we titrated down the fluid temperature to 33°C to mimic therapeutic hypothermia. We intended to perform the study from 32 to 40°C, but the water bath allowed only fluid temperature from 33 to 39°C. Moreover, we found that the insensible water loss was linearly correlated with this range of temperature change, a finding that was not addressed in previous publications.
The effect of the size of oxygenator on insensible water loss was not assessed in this study because Grist et al.6 demonstrated that the insensible water loss did not depend upon the surface area of the membrane oxygenator. Camacho et al.4 also demonstrated that insensible water loss was not affected by the membrane size. The effect of blood flow rate on the amount of insensible water loss was also not studied here because Camacho et al.4 have also shown that the pump flow rate will not affect the amount of insensible water loss.
As for the limitations of the present study, first, we did not have external validation of the sweep-gas flow rate and fluid temperature. Second, we used normal saline in the circuit and assumed that the effect of fluid loss will be the same as blood circulating in the ECMO circuit. Third, the question whether the membrane efficiency will be affected after a certain duration of ECMO circuit use is not addressed, and hence, it is possible to infer that the length of time the ECMO was put to use may affect the rate of fluid loss through the oxygenator membrane.
The amount of insensible water loss through adult ECMO circuit was linearly correlated to fluid temperature and sweep-gas flow rate. This study provides useful data for management of fluid balance in adult patients requiring ECMO support in the critical care setting.
1. Palder SB, Shaheen KW, Whittlesey GC, Nowlen TT, Kundu SK, Klein MD. Prolonged extracorporeal membrane oxygenation in sheep with a hollow-fiber oxygenator and a centrifugal pump. ASAIO Trans. 1988;34:820–822
2. Peek GJ, Mugford M, Tiruvoipati R, et al.CESAR Trial Collaboration. Efficacy and economic assessment of conventional ventilatory support versus
extracorporeal membrane oxygenation for severe adult respiratory failure (CESAR): A multicentre randomised controlled trial. Lancet. 2009;374:1351–1363
3. Lawson DS, Holt D. Insensible water loss
from the Jostra Quadrox D oxygenator: An in vitro
study. Perfusion. 2007;22:407–410
4. Camacho T, Totapally BR, Hultquist K, et al. Insensible water loss
during extracorporeal membrane oxygenation: An in vitro
study. ASAIO J. 2000;46:620–624
5. Alexander PJ, Lawson DS, Cornell J, Craig DM, Cheifetz IM. Insensible water loss
from the medtronic minimax oxygenator: An in vitro
study. ASAIO J. 2006;52:206–210
6. Grist G, Kangas B, Haney B, et al. Insensible water loss
from the Sci-Med 0.8M artificial lung during neonatal ECMO. 1998 4th
Annual CHNMC ECMO Symposium (Abstract)
7. Heulitt MJ, Pickert CBZwischenberger JB, Bartlett RHECMO. . ECLS and fluids, electrolytes and renal function. Extracorporeal Cardiopulmonary Support in Critical Care
. 1995 Ann Arbor, MI ELSO:123–135 pp.
8. Gill M, O’Shaughnessy K. Insensible water loss
from the Hilite 2400LT oxygenator: An in vitro
study. Perfusion. 2013;28:70–75
9. Yang M, Deng X, Laroche G, Hahn C, King MW, Guidoin RG. A capillary method to measure water transmission through polyurethane membranes. ASAIO J. 1997;43:890–896
10. Yu K, Long C, Hei F, et al. Clinical evaluation of two different extracorporeal membrane oxygenation systems: A single center report. Artif Organs. 2011;35:733–737
11. Formica F, Avalli L, Martino A, et al. Extracorporeal membrane oxygenation with a poly-methylpentene oxygenator (Quadrox D). The experience of a single Italian centre in adult patients with refractory cardiogenic shock. ASAIO J. 2008;54:89–94