How Much Sterile Saline Should be Used for Efficient Lavage During Total Knee Arthroplasty? Effects of Pulse Lavage Irrigation on Removal of Bone and Cement Debris

Materials and Methods 

Subjects comprised 8 patients with osteoarthritis who underwent primary TKA (LPS-flex, Zimmer, Warsaw, Ind) by the same surgeon. Mean patient age was 74 years (range, 47-82 years). All patients underwent complete resurfacing of the femur, tibia, and patella, and all components were cemented using PMMA bone cement (Zimmer). Before cementing components, 3 L of pulse lavage was routinely performed to remove bone debris and obtain reliable cement penetration 11, 12, 13. Care was taken to remove all excess PMMA and bone debris after cementing all components. Pulse lavage with 8 L of sterile saline was performed using a Pulsavac pulsatile jet irrigator (Zimmer). In all patients, a tourniquet was used during the surgery and remained inflated during 8 L of pulse lavage. Aspirated fluid was collected in a 1-L aliquot and was filtered through a nylon membrane filter with 1-μm pores. After NaOH digestion of collected specimens, the remaining undissolved debris was stained using alizarin red to distinguish between bone and PMMA debris. The debris then underwent Fourier transform infrared spectroscopy (FTS65A, Biorad, Cambridge, Mass) to identify materials within samples. Fourier transform infrared spectroscopy measures the absorption of various wavelengths of infrared light by the material of interest. Absorption spectra of samples were compared with standard spectra obtained from known materials (eg, bone, PMMA, etc) to determine the identities of the materials under analysis.

After alizarin red staining, stained and nonstained debris was examined under microscopy, and images were transmitted to a computer using image acquisition system. Image processing and determination of the number and size of cement and bone debris particles in a 1-L aliquot of fluid were performed using WinROOF version 5 software (Mitani, Fukui, Japan). Particle sizes were expressed using both absolute maximum diameter and equivalent circle diameter (ECD), which is the diameter of a circle having the same area as the measured sample of interest.

Results 

Debris isolated by NaOH digestion was clearly distinguished into red and white particles on plastic dishes by alizarin red staining (Fig. 1A). Fourier transform infrared spectroscopy analysis identified red particles as calcium phosphate (ie, bone) and white as PMMA (ie, bone cement) (Fig. 1B). No other materials were identified in the isolated debris. As assessed by image analysis software, numbers of bone and PMMA cement debris particles detected in 1-L lavage fluid peaked at first 1-L lavage and gradually decreased with increasing volume of lavage fluid (Fig. 2A). This was a common finding in all 8 patients. When debris was sorted for macroscopically visible particles (ECD >1 mm), a similar trend was observed (Fig. 2B). As the number of debris particles with ECD of more than 1 mm represented less than 5% of the total debris, data were slightly scattered, particularly for PMMA debris. When the number of debris particles detected in individual 1-L lavages was expressed as an average of 8 patients, significant differences were found between first vs second and third vs fourth lavages for bone, and between first vs second, second vs third, and third vs fourth lavages for PMMA (Fig. 2C). However, no significant difference was identified between fourth vs fifth lavage in either bone or PMMA. Consequently, both types of debris were efficiently removed up to the fourth lavage, reaching a plateau beyond the fourth lavage.

View Large Image Download to PowerPoint

Fig. 1. Identification of debris isolated from lavage fluid at the time of TKA. A, Representative photograph of debris isolated from 1 L of lavage fluid. Debris trapped on the nylon membrane filter was digested using NaOH, followed by alizarin red staining. Both stained and nonstained debris are present. B, Representative data of Fourier transform infrared spectroscopy analysis showing absorption spectra. Alizarin red–stained debris represents calcium phosphate (upper panel), whereas white nonstained debris represents PMMA (lower panel).

View Large Image Download to PowerPoint

Fig. 2. Bone (left panel) and PMMA cement debris (right panel) isolated from each 1-L lavage. A, Total debris from each of the 8 patients. B, Number of debris particles more than 1 mm in diameter for each patient. C, Mean number of total debris particles for the 8 patients. Data are expressed as mean ± SD.

Regarding size changes in debris removed during the course of 8-L lavage, mean ECD was basically constant for each 1-L lavage, whereas absolute maximum diameter peaked with the first lavage and gradually decreased with increasing volume of lavage fluid (Fig. 3). This discrepancy between average ECD and absolute maximum diameter is explained by the fact that particles with ECD of more than 1 mm accounted for less than 5% of the total debris (Fig. 1), and the number of macroscopically large particles was insufficient to affect mean ECD.

View Large Image Download to PowerPoint

Fig. 3. Sizes of bone and PMMA cement debris particles isolated from each 1-L lavage. A, Mean ECD for the 8 patients. B, Mean absolute maximum diameter for the 8 patients. Data are expressed as mean ± SD.

Discussion 

Recently, numerous investigators have reported that biologic responses to biomaterial particles are size-dependent. Horowitz et al has shown that PMMA particles with a diameter of 7 μm or less are phagocytosable by a single macrophage and capable of stimulating macrophages to produce bone-resorbing cytokines, whereas larger particles are stabilized after fibrous encapsulation [14]. Another study using an air-pouch model in rats revealed that small PMMA particles (<20 μm) induce greater inflammatory reactions than larger particles (50-350 μm), as determined by the release of tumor necrosis factor α, neutral metalloproteinase, and prostaglandin E2 [15]. Although tissue retrieval studies of failed prostheses have shown that particles resulting from biomaterial degradation are mostly submicrometer in size 16, 17, 18, the present results indicate that the diameter of particles generated intraoperatively is 200 to 240 μm for bone and 250 to 340 μm for PMMA, suggesting that particles generated intraoperatively would tend to induce third-body wear at articulating surfaces rather than inflammatory reactions at the bone-implant interface. In particular, when the amount of these particles exceeds the saturation level of macrophage phagocytosis or fibrous encapsulation, the remaining free particles within joint fluid would cause substantial third-body wear, leading to early polyethylene failure and subsequent aseptic loosening of the prostheses 19, 20.

Pulse lavage irrigation in TKA is now a popular technique in many hospitals for removing bone and PMMA debris before or after cementing implant and preventing potential bacterium infection. The effectiveness of pulse lavage irrigation has been reported 10, 11, 12, 13, but no previous studies have clarified the volume of lavage fluid suitable for efficient removal of bone and PMMA debris during TKA. The present study indicated that pulse lavage with 4 L of sterile saline is sufficient for removing both bone and PMMA debris intraoperatively. Considering bacteria as particles of approximately 1 μm diameter, 4 L of pulse lavage may be effective for removing not only bone and PMMA particles, but also bacterial particles.

Interestingly, the present results indicate that large debris tends to be removed earlier during the course of pulse lavage, whereas the mean ECD of removed debris remains constant. This is because only 5% of total debris is more than 1 mm in diameter, and this small amount of visible debris cannot affect mean ECD. Macroscopic examination by surgeons to determine whether pulse lavage is sufficient during an operation thus seems unlikely to be of any real use.

In conclusion, we have demonstrated that 4 L of pulse lavage irrigation is sufficient for removing cement and bone debris after cementing components of TKA. As a result, more than 82% of bone debris and more than 75% of cement debris can be removed with 4-L lavage. Although there has been no correlation between sufficient pulse lavage and any clinical outcome, we believe that making an effort to remove this debris intraoperatively may decrease the risk of third-body wear and subsequent aseptic loosening in the early years of TKA.