The influence of radio frequency ablation on intra-articular fluid temperature in the ankle joint - a cadaver study

Arthroscopic surgery of the ankle joint is widespread and tissue ablation is one of the main indications in the “Soccer’s Ankle” or anterior impingement. In this condition, fibrous tissue causes pain above the anterior aspect of the ankle joint and the removal of this tissue is an adequate therapy to provide pain-free motion. Most of these devices use electromagnetic energy for shrinking, coagulation or ablation of tissue [3, 4]. In contrast, we used a bipolar device in which the energy comes from a plasma layer at the tip of the wand. Radio frequency (RF) ablation devices are widely used to remove soft tissue in arthroscopic surgery. The basic idea of a RF device is to destroy soft tissue by electrolyte plasmarization with the side effect of increased heating of the irrigation fluid. To avoid intraarticular heat, the RF device has vents that are inset at the tip to increase the outflow of the arthroscopic irrigation fluid. Unfortunately, some hot water, gas and denatured material may escape into the joint cavity where it might lead to unwanted increased local temperatures. Recent publications have shown dermal burns of patients due to hot water spilling originating from the RF ablation process and there are also some cases of glenohumeral chondrolysis after labral repair in hip arthroscopy [1‐7]. Different reports investigated safety limits and found that chondrocyte damage may occur at temperatures as low as 45 °C/113 °F. Evaluations of the ability of the chondrocytes to recover showed a sharp increase of chondrocyte death between 50 to 55 °C (122 to 131 °F) [8, 9]. Thus, in recent publications on temperatures in shoulder joints [10] and hip joints [11], a 50 °C/122 °F criterion was used as limit for safe temperatures. Both publications report temperatures exceeding the critical point of 50 °C, which can easily be reached by RF ablation in the shoulder and hip joints depending on the flow of the irrigation fluid and their extraction. Until now, only limited data exists on the effects of RF ablation in the ankle joint, where especially the resection of hypertrophic synovia is useful and frequently used [12].

Anatomically, the capsular volume of the hip (2.5–10 ml) [13] is comparable to the volume of the ankle joint (6–10 ml) [14]. We therefore hypothesized that temperatures exceeding 50 °C/ (122 °F) may be reached by RF ablation. In our study, we wanted to investigate the impact of the irrigation flow rate, the controller unit voltage setting and the distance from the heat source on maximum temperatures, mean temperatures, time to reach the 50 °C/122 °F limit and – in view of the fact that the results of thermo-fluid dynamics experiments may vary substantially – the percentage of experimental runs exceeding 50 °C/122 °F with the same parameter settings.

A controlled laboratory study was designed using six formalin-fixed human cadaver ankle joints. All human specimen enrolled in this study were post-mortem donors to the Anatomical Institute of the University of Munich. The use of donated post-mortem specimens for scientific investigations is in accordance with the Declaration of Helsinki and was approved by the ethical committee of the University of Munich. Surrounding soft tissue around the ankle was preserved. The experiments were performed at room temperature using an antero-medial and antero-lateral portal. The irrigation inflow was placed in the antero-medial portal, while the bipolar RF device “Ambient Super Turbo Vac 90 IFS” with the Quantum II controller (Arthocare Corporation, Austin, Texas, USA) was placed via the antero-lateral portal (Fig. 1). The device uses a physical bipolar radio frequency process to stimulate electrolytes in the conductive natrium chloride solution. The energized particles in the plasma denaturize organic molecular bonds and dissolve tissue at temperatures between 40 °C and 70 °C/104 ° F and 158 °F, a current does not pass through the tissue. The result is volumetric removal of the target tissue with marginal collateral tissue damage. The RF device has vents at the tip, which are the outlets for the arthroscopic fluid. The Quantum II controller measures the temperature at the vents and cuts the power supply when temperatures exceed 40 °C/104 °F. Different voltage levels are possible, allowing a potential adaption according to the tissue treated.

All measurements were performed under direct arthroscopic vision, providing a free and unobstructed flow path for the irrigation fluid. To measure the temperature inside the ankle joint, three thermoprobes (TP) were placed 3, 5 and 10 mm off the tip (Fig. 2). Continuous temperature measurement was performed by a multichannel fiberoptical thermometer unit, which is sensitive enough to detect temperature changes of 0.01 °C. Temperatures were recorded and analyzed each second, while a measurement cycle took 120 s. In our view, the flow is the key parameter to excess heat in arthroscopy (Fig. 3). In order to vary the plasma power inside the cavity, the voltage was set from 1 to 5 and 9 at the end (minimum, medium and maximum level). The irrigation flow was controlled using the inlet pressures with 10, 25, 50 and 100 mmHg. The experiments started at room and irrigation fluid temperature of 20 °C/68 °F. During the 120 s, there is very little conductive heat flow from the cavity to the surrounding tissue and the heat capacity of the fluid in the ankle joint cavity is small (5–10 ml) (1). The experiments show that the heat escaping from the RF device leads to a heat build-up within 20 s at low coolant flow conditions. This clearly indicates that using 37 °C/98.6 °F instead of 20 °C/68 °F does not affect the maximum temperatures, but only reduces the time span before the stationary conditions are reached by a few seconds.

After having run four experiments for each combination as mentioned above, we analyzed our material and found that all experiments with the 100 mmHg pressure setting did not exceed the 50 °C/122 °F criterion, while the criterion was exceeded at least in one experiment for the other parameter combinations. However, we found no detectable effect of the voltage settings. Thus, the experiments with the same pressure setting are statistically in the same group, which means that we had 12 experiments for each pressure setting, which is sufficient for a meaningful statistical evaluation.

The purpose of this study was to investigate the effect of RF use in the ankle joint and how the three parameters irrigation fluid pressure, thermometer probe distance from the wand tip and the voltage setting of the controller affect the maximum temperatures in the ankle joint. The purpose was also to derive recommendations for the safe use of RF in the ankle joint.

There have been several reports about the deleterious effects of thermal damage to chondrocytes in shoulder arthroscopy resulting in chondrolysis [15, 16] and other papers indicate that a 50 °C/122 °F criterion should be observed since only temperatures below 50 °C/122 °F seem to be safe [9]. Two papers on the effect of RF use in the hip joint [11] and the shoulder joint [10] were published recently. Both papers found that temperatures in the joint, which exceed the criterion of 50 °C/122 °F, can be measured under no flow or low flow of irrigation fluid. From the analysis of the physical problem, it follows that the power output of the RF wand and the amount of irrigation flow are the key parameters, which influences the thermo-fluid dynamics. A typical phenomenon in thermo-fluid dynamics is temperature fluctuation. To measure these fluctuations close to the tip of the wand, three TPs were attached to the wand. From the physical analysis, it is also clear that there are many other parameters which may influence the temperature and which are not controlled in the experiment and therefore provide a random contribution to the data.

As the ankle joint volume (6–10 ml) is similar to the volume of the hip joint cavity (2.5–10 ml), we expected findings similar to those in the paper [10]. The regression analysis of the maximum temperatures (absolute maximum) and the mean temperatures (averaged over time) showed that 80% of the variances can be explained by the square root of the pressure (p < 0.001) and by the TP distance, which is also statistically significant (p < 0.05).

The effect of the voltage setting of the controller was random, which is hard to understand if one assumes that the voltage setting is relevant for the power output of the RF device. The maximum temperatures of the 12 runs at 100 mmHg were well below the 50 °C criterion, but for all other pressure settings, some or all runs clearly exceeded the criterion. It has to be concluded that high pressure has to be applied for safe temperatures. The physical analysis indicates that only the pressure difference across the cavity is relevant, thus an additional vacuum at the wand could be used to obtain the same thermo-fluid dynamics in the cavity with lower inlet pressure. The disadvantage of high pressures in the cavity is that water causes edema in the surrounding tissue. The situation can be improved by applying additional suction vacuum to reach the required pressure difference.

One option to reduce maximum temperatures in the ankle joint is the use of a turbulence-generating inflow opening at the arthroscope, which improves thermal mixing.

In addition, two issues were analyzed that might lead to even higher temperatures compared to those observed in this study: Hot non-condensable bubbles and outflow plugging. As the heat capacity of a non-condensable gas is lower than the heat capacity of water or tissue by orders of magnitude, it has to be concluded that hot non-condensable bubbles do not represent a potential hazard. As the RF device emits energized particles that break the molecular bounds within the tissue causing the tissue to dissolve, the ablation process generates more or less vapor and not particles, which might lead to plugging. Nevertheless, plugging cannot be ruled out as tissue may be cut loose and may be sucked into the RF device opening in the ablation process. Comparing the data found in our study and the studies on hip [10] and shoulder ablation [9], we could find similar temperature levels and no significant effect of the joint volume in view of the fact that the glenohumeral joint volume is more than twice the volume of the hip or the ankle joint [13, 14, 17].

Finally, the 50 °C/122 °F criterion is rather strict and does not reflect the chondrocyte damage mechanism. Obviously, a short contact with 50 °C will do no harm. Therefore, a better criterion that is based on more sophisticated and in-depth studies should be developed.

We deem the experimental settings of the study to be comparable to the surgical setting. We simulated the use of a RF device and were able to measure temperatures at the tip of the device directly. Although continuous use of the RF device for 60 s might be extreme, even a shorter time period (2 s) is able to cause cartilage damage. Therefore, it is tremendously important to ensure the correct irrigation and thereby diminish the risk of cartilage damage significantly.

In this context, it is crucial to keep in mind that not only high temperatures can have deleterious effects on the surrounding tissue, but also the use of increased water pressure. There have been reports about compartment syndromes of the lower leg after knee arthroscopy with injury to the posterior capsule [18] and of the anterior compartment after ankle arthroscopy in Maisonneuve fractures [19]. Mc Brayer et al. [20] could find an increase of swelling and deltoid muscle pressure during shoulder arthroscopy of 9 mmHg - particularly in operations lasting more than 90 min. Fortunately, they could not find a negative effect on the outcome in their study group.

Still, one has to think of the negative effects of increased pump pressures and what is more, Ross et al. [21] showed in their study that the operative field fluid pressure and the pressure readout might differ considerably.

There are limitations in this study: first, the use of cadaveric specimens leads to a different distraction of the ankle joint and the ankle joint cavity will differ in volume and shape affecting the flow pattern and fluctuations inside the cavity. Six cadaveric specimens cannot represent the full range of variability of human anatomy. Second, the procedure in this study with continuous ablation for more than 60 s may not reproduce the typical clinical scenario. Third, the cadaveric specimens were evaluated at room temperature and not at normal body temperature. Normal blood flow might work as a heat sink and dissipate heat. However, temperatures above the criterion of 50 °C/122 °F might be reached even faster at normal body temperature.

This study investigated the temperatures in the ankle joint during RF ablation. It showed that the key parameter to guaranty temperatures below the 50 °C/122 °F criterion is irrigation pressure. At pressures of 100 mmHg, temperatures in the ankle joint were found to remain well below this criterion. At lower pressure levels, temperatures clearly exceeding 50 °C/122 °F were reached in some or all experiments. Until a more sophisticated in-depth study reveals more detailed information on the effect of RF ablation on the temperatures in the ankle joint and potential hazard for chondrocytes, the authors recommend to keep the pressure difference across the ankle joint as high as feasible, the ablation time short and the temperature of the irrigation fluid low.