Comparison of Four Different Pediatric 10F Aortic Cannulae During Pulsatile Versus Nonpulsatile Perfusion in a Simulated Neonatal Model of Cardiopulmonary Bypass

Abstract

We compared four commercially available 10F pediatric aortic cannulae with different geometric designs (DLP—Long tip, DLP—Short tip, RMI—Long tip, and Surgimedics—Short tip) during pulsatile versus nonpulsatile perfusion in terms of pressure drops and surplus hemodynamic energy (SHE) levels in an in vitro neonatal model of cardiopulmonary bypass. The pseudo patient was subjected to seven pump flow rates at 100 ml/min increments in the 400–1,000 ml/min range. A total of 44 experiments (n = 22, nonpulsatile; n = 22, pulsatile) were performed at each of the seven flow rates. Surgimedics had significantly higher pressure drops than the other three cannulae at various flow rates during nonpulsatile and pulsatile perfusion, respectively. When the perfusion mode was changed from nonpulsatile to pulsatile flow, SHE levels at both precannula and postcannula sites increased seven to nine times at all flow rates in all four cannulae. Surgimedics generated a significant lower SHE level when compared with the other three cannulae at all flow rates at both precannula and postcannula sites. The results suggest that different geometries of aortic cannulae have a significant impact on pressure drops of the cannulae as well as hemodynamic energy generation and delivery. Pulsatile perfusion generates more “extra” hemodynamic energy when compared with the nonpulsatile perfusion mode with all four cannulae used in this study.

The advantages of pulsatile over nonpulsatile perfusion in terms of hemodynamic energy delivery have been the focus of several studies in our laboratory during the past decade.^1–5 We now have conclusive evidence from these studies showing that pulsatile flow delivers greater hemodynamic energy levels^6–8; however, it is now important to realize that the components of the cardiopulmonary bypass (CPB) circuit, especially the pump, oxygenator, and aortic cannula, have a drastic effect on the amount of this beneficial hemodynamic energy delivered to the patient.

In our previous studies, we documented the performance of the different heart-lung machines with Food and Drug Administration approved⁹ and different pediatric oxygenators with pulsatile flow versus nonpulsatile flow on in vitro models¹⁰ as well as on in vivo models.¹¹ However, to generate adequate pulsatility and deliver sufficient hemodynamic energy to the patient during CPB, the aortic cannula, which is one of the most critical components of CPB, should be chosen carefully during pulsatile perfusion. Although the sizes of the cannulae are the same, the length and geometry may be different. The tip of the cannula is either hard plastic or a reinforced wire. Tip dimensions are commonly stated in French (F) [1F equals an outer diameter (OD) of 0.33 mm]; however, inner diameter (ID) may differ significantly (Table 1).

Table 1: Inside and Outside Diameters of Four Different 10F Aortic Cannulae

A 10F cannula is typically used during pediatric CPB for infants. The objectives of this investigation were to 1) compare four different 10F commercially available pediatric aortic cannulae with different tips and geometry in terms of pressure drop during normothermic CPB in a simulated neonatal model; and 2) evaluate pulsatile and nonpulsatile perfusion modes for each cannula in terms of surplus hemodynamic energy (SHE) levels.

Materials and Methods

Experimental Setup

The experimental setup in this simulated model is composed of three parts.

The first part is the extracorporeal circuit, which consists of a Jostra HL-20 heart-lung machine (Jostra USA, Austin, TX) for both pulsatile and nonpulsatile modes of perfusion, a Jostra-30 heat-cooler system (Jostra USA), a Capiox Baby RX hollow-fiber membrane oxygenator (Terumo Corporation, Tokyo, Japan), a Capiox pediatric arterial filter (Terumo Corporation), 5 ft of arterial tubing, and 6 feet of venous tubing of the COBE Heart/Lung Perfusion Pack (COBE Cardiovascular Inc, Arvada) for pediatric patients with a ¼ in diameter.

The second part is the simulated Penn State patient, which includes one of four arterial pediatric cannulae: DLP 77010—Long tip (Medtronic, Inc, Minneapolis, MN); DLP 75010—Short tip (Medtronic, Inc); RMI Fem II-010-A—Long tip (Edwards Life sciences LLC, Irvine, CA); and Surgimedics TMP—Short tip (Texas Medical Products, Inc, TX) (Figure 1). Aortic compliance, systemic resistance, and venous compliance were attached to the extracorporeal circuit. A 70 cc Penn State pneumatic left ventricular assist device with valves removed was employed as the aortic compliance. To provide an adequate aortic compliance, the air side of the pneumatic pump was connected to a 0.3 L glass flask in which the mean arterial pressure (MAP) was regulated to approximately 40 mm Hg. A Hoffman clamp, positioned between the aortic and the venous compliance chambers provided the peripheral resistance.

Figure 1.: Picture of four different 10F aortic cannulae.

The third part is the data acquisition system. The pump flows and pressure waveforms were recorded at the precannula and postcannula sites. Maxxim disposable pressure transducers model 041500503A (Maxxim Medical, Inc., Athens, TX) were used for pressure measurements and a Transonic ultrasound flowsensor model 6XL (Transonic System Inc., Ithaca, NY) attached to the Transonic flowmeter model TS410, was used for flow measurements. Data were acquired using a custom made LabView software program (National Instruments, Austin, TX) with a sampling rate of 1,000 samples/s. Approximately 20 seconds of data were recorded at each test condition. Mean circuit pressure, energy equivalent pressure (EEP), and SHE were calculated from 20 seconds of data.

The whole mock loop was primed with a lactated Ringer’s solution. The total priming volume was 550 ml, including the extracorporeal circuit (350 ml) and Penn State neonatal patient (200 ml). A temperature probe was placed on the venous line close to the reservoir for continually monitoring the temperature of the mock loop. The circuit temperature was maintained around 35°C by constantly adjusting the temperature of the heat-cooler unit.

Pulsatile Mode Setting

Pulsatile flow settings in this roller pump are 10% of the base flow, 120 beats per minute of the pump rate, 20% of the pump head start point, and 80% of the pump head stop point. The pump start and pump stop timing points reflect the time between two R waves of an electrocardiogram, and the pump start and pump stop points are set as a percentage of every circle.

Experimental Design

The pseudo patient is subjected to seven pump flow rates 100 ml/min increments in the 400–1,000 ml/min range. At each pump flow rate, the MAP is set at a constant 40 mm Hg via Hoffman clamp. When the target pump flow rate is achieved, a 20 second segment of the pressure and flow waveforms with nonpulsatile flow are recorded. The perfusion mode is then switched to pulsatile flow. Experiments were divided into eight groups, depending on the types of perfusion modes (pulsatile vs. nonpulsatile) or the types of cannulae. A total of 44 experiments were performed (n = 22 with nonpulsatile and n = 22 with pulsatile) at each of the seven flow rates: group A-NP: DLP 77010 with nonpulsatile flow, (n = 6); group A-P: DLP 77010 with pulsatile flow, (n = 6); group B-NP: DLP 75010 with nonpulsatile flow, (n = 5); group B-P: DLP 75010 with pulsatile flow, (n = 5); group C-NP: RMI with nonpulsatile flow, (n = 5); group C-P: RMI with pulsatile flow, (n = 5); group D-NP: surgimedics with nonpulsatile flow, (n = 6); group D-P: surgimedics with pulsatile flow, (n = 6).

Pressure Drop of the Cannulae

Pressure drops of the cannula were calculated for the all the cannulae with two perfusion modes:

Surplus Hemodynamic Energy

At both experimental sites (precannula and postcannula), SHE was calculated by multiplying the difference between the EEP and the mean pressure (MP) by 1,332:

where f is the flow and p is the pressure.¹²

Statistical Analysis

For each flow rate and stage, a linear mixed-effects model was fit to the data to assess differences in SHE, mean pressure, and cannulae pressure drop between cannulae and perfusion methods.¹³ The linear mixed-effects model accounts for the correlation among repeated measurements per simulated patient. The between-subjects factor for the model was cannula (DLP77010 long tip, DL75010 short tip, RMI long tip, Surgi short tip) and the within-subjects factor for the model was perfusion method (nonpulsatile, pulsatile). p values and 95% confidence intervals for mean difference estimates were adjusted for multiple comparison testing using Bonferroni’s procedure. All analyses were performed using the SAS software package (SAS Institute Inc., Cary, NC).

Results

Mean Circuit Pressure

Mean circuit pressures in pulsatile flow were slightly higher than nonpulsatile flow at each flow rate (400 ml/min–1,000 ml/min) with all four cannulae at precannula site. Regardless of perfusion modes, Surgimedics had significantly higher mean circuit pressure than the other three at precannula site at all flow rates (because postcannula pressure was set at a constant 40 mm Hg via Hoffman clamp for all cannulae). The mean circuit pressures were similar among DLP (long tip), DLP (short tip), and RMI (long tip) at precannula site at each flow rate with both pulsatile and nonpulsatile perfusion modes. Details are presented in Table 2 and Figures 2–5.

Table 2: Precannulae Mean Circuit Pressure Results (mm Hg)

Figure 2.: Waveforms of the precannula pressure, postcannula pressure, and pump flow with a rate of 600 ml/min during nonpulsatile and pulsatile perfusion with a DLP (long tip) cannula.

Figure 3.: Waveforms of the precannula pressure, postcannula pressure, and pump flow with a rate of 600 ml/min during nonpulsatile and pulsatile perfusion with a DLP (short tip) cannula.

Figure 4.: Waveforms of the precannula pressure, postcannula pressure, and pump flow with a rate of 600 ml/min during nonpulsatile and pulsatile perfusion with a RMI (long tip) cannula.

Figure 5.: Waveforms of the precannula pressure, postcannula pressure, and pump flow with a rate of 600 ml/min during nonpulsatile and pulsatile perfusion with a Surgimedics (short tip) cannula.

Pressure Drop of the Cannulae

Regardless of the perfusion modes, the pressure drops in the Surgimedics cannula were significantly higher than the other three cannulae at each flow rate, and the other three cannulae had very similar results. As the flow rate increased from 400 ml/min to 1,000 ml/min, the pressure drops increased in all of the cannulae with either nonpulsatile or pulsatile flow. When the perfusion mode was switched to pulsatile flow from nonpulsatile flow, pressure drops in pulsatile flow were slightly higher than in nonpulsatile flow. Details are presented in Table 3.

Table 3: Results—Cannula Pressure Drops (Precannula − Postcannula) (mm Hg)

SHE Levels

SHE levels at precannula site were similar among all four cannulae at all flow rates with nonpulsatile flow. Surgimedics generated significant lower SHE levels than the other three cannulae at the postcannula site at all flow rates with nonpulsatile flow. With the flow rate increasing from 400 ml/min to 1,000 ml/min, SHE levels decreased at the postcannula site in all of the cannulae with either nonpulsatile or pulsatile flow. When the perfusion mode was changed from nonpulsatile to pulsatile flow, SHE levels at both precannula and postcannula sites increased sevenfold to ninefold at all flow rates in all four cannulae. Surgimedics had a significantly lower SHE level when compared with the other three cannulae at all flow rates on both precannula and postcannula sites. Details are presented in Tables 4 and 5.

Table 4: Precannulae Surplus Hemodynamic Energy Results (ergs/cm³)

Table 5: Postcannulae Surplus Hemodynamic Energy Results (ergs/cm³)

Discussion

The aortic cannula is one of the most critical components of the extracorporeal circuit. A high blood flow rate through the narrow lumen of the tip may lead to a high pressure drop, high local velocities, and turbulence, thereby causing hemolysis,¹⁴ as well as significant correlations with the systemic inflammatory response syndrome during CPB. Particularly in neonates and infants, the pump flow rates are routinely maintained at 150–200 ml·kg⁻¹·min⁻¹ compared with only 50–80 ml·kg⁻¹·min⁻¹ in adult CPB cases. In addition, the pressure drops are a more important issue for the high-risk pediatric patients who must undergo a longer duration of CPB procedure.

In this study, regardless of perfusion mode, pressure drops of Surgimedics were significantly higher than the other three cannulae at each flow rate. Theoretically, cannulae that have larger ID and shorter length will present the lowest pressure drop for the same flow rate.¹⁵ According to Poiseuille law: Q=πR4ΔP/8ηL, where Q is the flow rate, η is the viscosity of the priming in the circuit, R is 1/2 ID of tip, which acts to the fourth power; whereas the ΔP (pressure drop) is linearly proportional with the L (length of the tip). Based on the measurements of the cannulae in Table 1, the inner and ODs varied in all four 10F size cannulae from three separate companies, even in two cannulae from the same company. Surgimedics has the smallest ID. Therefore, it is not surprising that Surgimedics generated the highest pressure drops in all the cannulae.

When the perfusion modes were changed from nonpulsatile to pulsatile, we found the mean circuit pressures in pulsatile flow slightly higher than nonpulsatile flow at the precannula site at each flow rate; we also observed pressure drops in pulsatile flow were slightly higher than in nonpulsatile flow in these four cannulae at each flow rate. However, changing perfusion modes significantly increased the SHE levels (up to 7–8 times compared with the baseline) in all four cannulae at both experimental sites (precannula and postcannula) with each flow rate. It has been documented that SHE is the “extra” energy generated only during pulsatile perfusion. There is considerable evidence showing a significant difference in SHE levels when comparing the two perfusion modes: when using the pulsatile flow, vital organ blood flow (brain, renal and myocardial) is significantly better maintained,^16–19 and the systemic inflammatory response syndrome is significantly reduced.²⁰

Furthermore, from SHE level results, we also found that Surgimedics produced the significantly lowest hemodynamic energy among all four cannulae at both experimental sites (precannula and postcannula) at each flow rate. Apparently, there is a significant correlation between pressure drops and hemodynamic energy delivery. We believe that both ID and length of cannula tip greatly impact hemodynamic energy level delivery. This correlates well with our previous results, which revealed that a long-tip cannula significantly damped pulsatility when compared with a short-tip cannula and demonstrated that the same type of aortic cannula with a slightly different ID size can make an enormous difference in the waveform characteristics.¹¹ Therefore, ID and OD, wall thickness, flexibility, and length of the tip of the aortic cannulae are important factors that need to be taken into consideration in aortic cannula design.

In this study, we used lactated Ringer’s solution to prime the extracorporeal circuit because all cannulae manufacturers publish cannula pressure drop results using water as a priming solution. In a clinical setting, however, the viscosity of the priming solution including blood will be much different than just using lactated Ringer’s solution or water. Further studies for comparing several arterial and venous cannulae with different perfusion modes using bovine blood are planned in our laboratory.

Conclusions

Surgimedics had significantly higher pressure drops than the other three cannulae at all flow rates during both nonpulsatile and pulsatile perfusion. Regardless of the type of cannulae, when the perfusion modes were changed from nonpulsatile flow to pulsatile flow, SHE levels at both precannula and postcannula sites increased at each flow rate. Surgimedics generated a significantly lower SHE level when compared with the other three cannulae at all flow rates on both precannulae and postcannula sites. The results suggest that pulsatile perfusion generates more “extra” hemodynamic energy when compared with the nonpulsatile perfusion mode with all four cannulae used in this study. Furthermore, the ID and geometry of the cannula also have a great impact on the amount of SHE delivered to the pseudo patient. Further research on a new design of the pediatric aortic cannula is warranted.