Longevity and Catheter-Related Infection Rates in Nonhuman Primates with Chronic Indwelling Intravenous Catheters
In preclinical models, indwelling intravenous catheters and vascular access ports are often essential components of biomedical research aimed at modeling human disease. For instance, animal models of drug self-administration are used for many reasons, including to assess abuse liability, to study physiologic and neurologic consequences of drug exposure, and to examine the efficacy of behavioral and/or pharmacological interventions. The most frequent route of drug self-administration in preclinical animal models is the intravenous route via indwelling intravenous catheters. The present study examined 23 years of drug self-administration studies in Old World macaques used in drug self-administration studies at Wake Forest University School of Medicine. The medical records for individually or pair-housed adult rhesus monkeys (n = 10 females and 172 males) and socially housed cynomolgus monkeys (n = 64 females and 92 males), all implanted with indwelling intravenous catheters and associated vascular access ports, were examined. The most frequent vein catheterized was the femoral vein, followed by the internal and external jugular vein; the least frequent was the brachial vein. The infection rates over 23 years and >500 catheters in cynomolgus and rhesus monkeys were 13.7% and 10.3%, respectively. The average catheter remained patent and implanted in the vein for 22.5 months in cynomolgus monkeys and 15.5 months in rhesus monkeys. These findings highlight significant strengths in using Old World macaques, both rhesus and cynomolgus, in long-term, longitudinal studies involving indwelling intravenous catheters.
Introduction
In many preclinical studies, chronic indwelling intravenous catheters and vascular access ports (VAPs) are used to administer drugs. One particular research field, animal models of substance use disorders, use chronically indwelling intravenous catheters and VAPs for drug self-administration experiments that involve the long-term maintenance and repeated use of this system.1,2 These catheters allow for chronic intravenous drug delivery under various schedules of reinforcement to assess the reinforcing effects of numerous drugs.1,3,4 Nonhuman primates (NHPs) are one animal model used to study drug self-administration. Although monkeys only make up a small percentage of animals used in preclinical research, their impact on human health is enormous.5,6 For instance, because NHPs are neuroanatomically, phylogenetically, cognitively, and socially similar to humans7 they provide the most translational model of drug self-administration in the context of substance use disorders.8,9 For drug self-administration studies, the use and maintenance of chronically indwelling intravenous catheters is crucial for success.9
Although surgical procedures for the implantation of these catheters have been published for >50 years, procedures for implanting and maintaining these catheters have markedly improved over time.8–11 Despite decades of indwelling catheter use in NHPs, there are still gaps in the literature concerning the longevity and sterility of these catheters in NHPs self-administering drugs of misuse. NHPs, unlike rodents, have large veins that allow multiple locations for catheter implantation. In rhesus and cynomolgus monkeys, the internal jugular, external jugular, femoral, and occasionally brachial and saphenous veins can be used for catheter implants.12 While larger veins and more viable vessels significantly extend how long a monkey can be used in drug self-administration studies relative to other animal models, there are significant challenges associated with optimizing catheter longevity.
One such challenge is bacterial infection. In the human population, peripheral intravenous catheters and VAPs are commonly implanted in healthcare settings. In fact, up to 60% of hospital inpatients receive at least one during their admission period.13 In this setting, the incidence of catheter-related bloodstream infections is relatively low, ∼0.1% or 0.5 per 1,000 catheter days.14 The incidence of infection has been shown to be higher in NHPs used in biomedical research. Several studies have examined the management of VAPs in NHPs and found that although 12% of VAPs were removed due to complications, only 5.1% of these were removed due to infection.10,11 In terms of catheter-related bloodstream infections, one study conducted in 1998 in rhesus monkeys found that across the span of 21 months, 30.2% of monkeys with catheters were diagnosed with catheter tract infections.15 Importantly, in that study, no VAPs were used and catheters extended through the skin and were accessed outside the body, which may have contributed to an increased risk of infection.
While the placement and maintenance of catheters in NHPs follows procedures similar to those used in humans, the housing conditions of these animals makes the long-term sterility of these catheters significantly more difficult. For example, monkeys in biomedical research are commonly housed in cages where both feces and urine are present despite daily cleaning by husbandry.16 Our laboratories study drug self-administration in socially housed monkeys that allows physical contact with cage mates.17 Taken together, these factors increase the risk of contamination of an intravenous catheter via introduction of bacteria from the hands and bodies of the monkey and cage mates.
Because catheter-related infections can reduce the monkey’s duration in a research study that requires a patent intravenous catheter and can potentially have negative consequences on animal health, understanding and reducing rates of catheter tract infection in monkeys self-administering drugs is critical.15 Such knowledge will not only maximize the health of monkeys in biomedical research but also prevent the premature removal of intravenous catheters. Thus, this study aimed to determine the rate of catheter-related infections, as well as bloodstream infections, and catheter longevity in socially housed male and female cynomolgus macaques and individually housed male rhesus monkeys across >20 years in a laboratory using NHP models of intravenous cocaine use. We chose the following primary dependent variables, which we thought were most relevant to assessing catheter-related incidences in macaques: sex of the monkey, with the hypothesis that catheters would remain patent longer in males due to their size; location of the catheter, with the hypothesis that femoral vein catheters would remain patent longest because of the size of the vessel; and number of repairs, with the hypothesis that the more repairs needed, the more likely the catheter would be removed due to infection. For these studies, patency was typically determined by drug self-administration behavior; occasionally we would attempt to draw blood from the VAP, but this is not a precise indication of catheter patency. We did not correlate age with any of these outcomes because catheters were implanted in the following order: (right, then left) femoral vein; internal jugular; external jugular; and brachial. As a result, monkeys were always older when they lost an external jugular or brachial catheter.
Materials and Methods
Subjects.
Data were used from 156 cynomolgus monkeys (Macaca fascicularis; 64 females and 92 males) and 182 rhesus monkeys (Macaca mulatta; 10 females and 172 males). Cynomolgus monkeys were housed in same-sex social groups of 4 while rhesus monkeys were either individually housed or housed in pairs. Numerous studies examining the reinforcing effects of various drugs of misuse such as cocaine and the efficacy of novel therapeutics have been published using these monkeys.18–23
Monkeys were housed in stainless-steel cages, typically 0.71 × 1.73 × 1.83 m (Allentown Caging Equipment, Allentown, NJ); for the socially housed cynomolgus monkeys, these cages had removable wire mesh partitions that separated monkeys into quadrants (0.71 × 0.84 × 0.84 m). The housing rooms were temperature (∼73 °F [23 °C]) and humidity (∼34%) controlled, and were on a 14-hour light/10-hour dark cycle, with the lights on between 6:00 am and 8:00 pm. All animals were fed standard laboratory chow (LabDiet 0549; Purina, St. Louis, MO) and maintained at healthy body mass, as determined by periodic weighing and visual inspection by laboratory staff and veterinarians. Enrichment was provided daily in the form of fresh fruits and vegetables, and water was always available in the home cage. Additional environmental enrichment was provided as determined by the IACUC of Wake Forest University Non-Human Primate Environmental Enrichment Plan, which includes use of chew toys, music, mirrors, and foraging feeders. Animal housing, handling, and experimental protocols were carried out according to the National Research Council Guidelines for the Care and Use of Mammals in Neuroscience and Behavioral Research and were approved by the Wake Forest University IACUC.
Catheter implantation.
For all catheter implantation surgeries, a chronically indwelling intravenous catheter (Hydrocoat catheter, 5 French; Access Technologies, Skokie, IL) and subcutaneous VAP (catalog no. CP4AC-5H for rhesus and no. CP6AC-5H for cynomolgus monkeys; Access Technologies, Skokie, IL) were implanted under aseptic conditions. Before each surgery, the monkey received the antibiotic cefazolin (30 mg/kg Kefzol IM; Marsam Pharmaceuticals, Cherry Hill, NJ). After inducing anesthesia with ketamine (10-15 mg/kg IM), or a ketamine (5 mg/kg)/dexmedetomidine (0.04 mg/kg) combination, anesthesia was maintained with 1.5% isoflurane gas for the remainder of the procedure. Monkeys were shaved around the incision site above the vein and on the back, where the VAP would eventually be located. The skin was prepared by 3 sets of scrubs alternating between a povidone-iodine scrub and 70% isopropanol. A final skin preparation was performed before an incision was made with a DuraPrep surgical solution (Solventum, Maplewood, MN). The proximal end of the catheter was inserted into a major vein and passed to the level of the inferior vena cava; the catheter was secured to the vein with 3.0 silk braided suture. After an incision was made off the midline of the monkey’s back, a subcutaneous pocket was created for the VAP via blunt dissection. The distal end of the catheter was subcutaneously threaded to this incision site using a rod with a slightly pointed end to tunnel just under the skin to the newly formed pocket, and the catheter was connected to the VAP. The catheter was secured to the VAP using a 3.0 polypropylene monofilament nonabsorbable suture, and the incisions were closed via 3.0 monofilament poliglecaprone 25 synthetic absorbable suture. Postoperatively, monkeys sedated with the ketamine/dexmedetomidine combination were given atipamezole HCl (0.2 mg/kg). From 2016 to 2023, monkeys received meloxicam (0.2 mg/kg IM prior to the surgery and then 0.1 mg/kg PO for 3-5 days following surgery) and ceftiofur sodium (2.2 mg/kg IM) for 7-14 days. Prior to 2016, monkeys received ketoprofen (5.0 mg/kg) for 3 days postoperatively and ceftiofur sodium (2.2 mg/kg IM) for 7-14 days. After drug self-administration sessions, each port and catheter were filled with ∼3.0 mL solution of heparinized saline (100 U/mL) to prolong patency and prevent clotting. This volume was determined to fill the VAP and catheter line.
Catheter maintenance procedure.
All monkeys were fitted with aluminum collars and trained to sit in a standard primate chair (Primate Products, Redwood City, CA). Behavioral experiments using drug self-administration were conducted in ventilated, sound-attenuating primate chambers (1.5 × 0.76 × 0.76 m; Med Associates, St. Albans, VT). Prior to each session (5-7 days per week), the area on the monkey’s back containing the VAP was prepared with an antiseptic povidone-iodine scrub (Medline Industries, Mundelein, IL) followed by isopropyl alcohol (Fisher Scientific, Fair Lawn, NJ). The area was given a final preparation with povidone-iodine scrub, and 5 minutes later, a 22-gauge Huber needle (Access Technologies, Skokie, IL) was inserted into the monkey’s port, connecting the catheter to an infusion pump. This peristaltic infusion pump (Cole-Parmer Instrument, Niles, IL) was located on top of the chamber for intravenous drug delivery at a rate of ∼1.5 mL per 10 seconds. Before each session, the pump was operated for ∼3 seconds to fill each monkey’s port with saline or the concentration of drug available for that session. Importantly for the assessment of catheter longevity, after use in a monkey, the Huber needle was cleaned by soaking in chlorhexidine solution (Aspen Veterinary Resources, Liberty, MO) and carefully flushed with water, dried and placed in an autoclave pouch, gas sterilized, and used again. This practice occurred until the needle became dull or damaged.
Occasionally, a monkey would require a VAP repair due to loss of integrity of the skin overlying the device. Anesthesia and postoperative recovery were identical between this procedure and the catheter implant procedure described above. For these repairs, a small incision was made along the existing VAP, and the VAP was removed via blunt dissection. After debridement, a new VAP was connected to distal end of the catheter and a new subcutaneous pocket was created for the VAP via blunt dissection. If a catheter and VAP had to be removed due to infection, ceftiofur sodium treatment was initiated and continued (pending culture results) for 2 weeks. Typically, catheter removal consisted of making an incision over the site where the catheter was secured to the vein. A suture was placed around the vein and catheter and the sutures securing the catheter were carefully removed. The catheter was then pulled out from the back of the monkey, at the site of the VAP. A new catheter, in a different vein, was typically not surgically implanted for at least 2 weeks after removal and until veterinarians cleared the monkey for a new catheter. The removed VAP was discarded, and a new VAP was implanted with the new catheter.
Data collection procedures and measures.
This study used data from longitudinal animal records that document all surgical procedures in individual monkeys as part of NIH-funded research at Wake Forest University School of Medicine. Instances of catheter tract infections, the number of repairs needed, and the length of time a catheter remained implanted in each monkey were documented according to catheter implantation site (femoral, internal jugular, external jugular, brachial, or saphenous vein). Each of the 328 monkeys included in this survey had a history of several catheter implantations; 300 total catheters were assessed in cynomolgus monkeys (99 female and 201 male) and 240 were assessed in rhesus monkeys (21 female and 219 male). Average weights for rhesus monkeys ranged from 6 to 13 kg and ranged from 2.5 to 10 kg in cynomolgus monkeys. For the cynomolgus monkeys, 37 of 300 catheters were currently implanted at the time data collection was completed. In the rhesus monkeys, 4 of 240 catheters were currently implanted. Catheter implants were excluded from analyses if no implantation or removal date was documented. Moreover, since some study outcomes were terminal, catheters were not included when they were implanted <3 months prior to euthanasia. In the early research involving rhesus monkeys, animals lived in their experimental chambers; these monkeys were excluded from these analyses because the procedures for maintenance of these catheters differed from the monkeys who were implanted with VAPs. Based on these criteria, a total of 43 catheter implants were excluded in the cynomolgus monkeys and 139 were excluded in the rhesus monkeys.
In animal records, laboratory staff had recorded suspected infections and confirmed infections. Suspected infections were based on visual inspection of the VAP and/or catheter implantation site. If swelling or irritation was observed, laboratory staff opted to remove the catheter before there were clinical signs of infection such as an abscess at the implant site or pus around the VAP. In the case of clear signs of infection (ie, evidence of purulence, incision site red and/or open, poor appetite, lethargy, elevated body temperature) veterinary staff was alerted, blood cultures were taken, and the catheter was removed. Positive cultures were documented as confirmed catheter-related bloodstream infections. Although the reasons a catheter was removed were not always documented, if the catheter did not have a suspected or confirmed infection, 2 frequently reported reasons were catheter occlusion and leakage from the insertion site.
Data analysis.
The primary dependent variables for this study were the number of confirmed and suspected catheter infections and catheter longevity (number of months that a catheter remained viable). Statistical analyses were conducted separately for cynomolgus monkeys and rhesus monkeys. Descriptive statistics were used to evaluate the overall rate of confirmed infections and suspected infections across 23 years. Bivariate correlations were determined to test whether there was a relationship between the number of confirmed or suspected infections and the year in which the catheter was implanted. Furthermore, Poisson regression analyses were conducted to determine whether the sex of the monkey (in cynomolgus monkeys) or implantation site of the catheter (femoral, internal jugular, external jugular, brachial, or saphenous vein) predicted the number of confirmed and suspected catheter infections across 23 years. In rhesus monkeys, an analysis on sex was not conducted due to a low number of female monkeys. Poisson regression analyses were also conducted to determine whether the number of VAP repairs or the longevity of the catheter predicted whether a catheter developed a confirmed or suspected infection.
To evaluate catheter longevity, a univariate ANOVA was conducted to determine whether catheter longevity varied across implantation sites (femoral, internal jugular, external jugular, brachial vein). Because only one catheter was implanted in the saphenous vein, this data point was excluded for the ANOVA. A significant ANOVA was followed by pairwise multiple comparisons (Holm-Sidak) post hoc tests. An independent samples t test was also conducted on the data from the cynomolgus monkeys to evaluate whether catheter longevity varied based on the sex of the monkey.
Results
Catheter infection rate.
In cynomolgus monkeys, 300 catheters were assessed. Femoral vein implants accounted for 180 of these catheters, 69 were internal jugular implants, 37 were external jugular implants, 13 were brachial implants, and 1 was a saphenous implant (Figure 1A). In the cohort of cynomolgus monkeys, 13.7% (41/300) of all catheters developed a confirmed infection and 10.3% (31/300) were removed due to suspected infection throughout the 23-year period (Figure 2A). There was no relationship between the year the catheter was implanted or the frequency of confirmed infections and suspected infections (P > 0.05). In fact, the frequency of confirmed and suspected infections remained stable and low across the 23-year period (Figure 3A).
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-25-073
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-25-073
In rhesus monkeys, 240 catheters were assessed; 157 were femoral implants, 47 were internal jugular implants, 25 were external jugular implants, and 11 were brachial implants (Figure 1B). In these rhesus monkeys, 10% (24/240) of all catheters developed a confirmed infection, and 4.2% (10/240) of all catheters were removed due to a suspected infection throughout the 23-year period (Figure 2B). In rhesus monkeys, there was also no relationship between the year the catheter was implanted or the frequency of confirmed infections and suspected infections (P > 0.05). This frequency remained low and stable across 23 years (Figure 3B).
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-25-073
In the cynomolgus monkeys, a Poisson regression demonstrated that the number of confirmed infections (Table 1, open cells) was predicted by the sex of the monkey such that males were 4.56-fold (95% CI, 1.62-12.87) more likely to develop a confirmed infection when compared with females (χ2(1) = 8.30, P = 0.004). In fact, out of 41 confirmed infections, 90.24% (37/41) occurred in male monkeys (Figure 4). This relationship was not present (P = 0.639) for the suspected infections (Table 1, shaded cells). There was no relationship (Table 1) between the location of the implant and the frequency of confirmed (P = 0.436) or suspected (P = 0.393) catheter infections (but see catheter longevity analyses, below). The number of VAP repairs was significantly related to the frequency of confirmed infections such that a higher number of repairs was related to a higher frequency of confirmed infections (χ2(1) = 6.881, P = 0.009). For every additional repair, the risk of having a confirmed infection increased by 13% (95% CI, 3.2%-24%) (Table 1, open cells). This relationship was trending with suspected infections (χ2(1) = 2.99, P = 0.083) such that for every additional repair, the relative risk of suspected infection increased 10% (CI, −1.3% to 24%) (Table 1, shaded cells). There was also a trending relationship such that the frequency of confirmed catheter tract infections increased the longer the catheter was implanted (χ2(1) = 3.61, P = 0.058). There was a 1% (95% CI, 0%-2%) increase in the relative risk of infection for each extra month that a catheter was implanted into a vein (Table 1, open cells). This increase in relative risk of infection was not significant when the frequency of suspected catheters was used as the dependent variable (P = 0.505) (Table 1, shaded cells). For all models, a Pearson χ2 statistic demonstrated that the data were not overdispersed (χ2/df < 1).
| Independent variable | β | SEM | Odds ratio: Exp(B) | 95% CI | P value |
|---|---|---|---|---|---|
| Sex | 1.52 | 0.52 | 4.56 | 1.62-12.78 | 0.004 |
| 0.186 | 0.39 | 1.20 | 0.55-2.62 | 0.639 | |
| Location of implant | −0.151 | 0.19 | 0.860 | 0.59-1.26 | 0.436 |
| −0.195 | 0.23 | 0.823 | 0.53-1.29 | 0.393 | |
| Repairs | 0.124 | 0.05 | 1.13 | 1.03-1.24 | 0.009 |
| 0.10 | 0.06 | 1.10 | 0.987-1.24 | 0.083 | |
| Longevity | 0.009 | 0.0049 | 1.01 | 1.00-1.02 | 0.058 |
| −0.006 | 0.0083 | 0.99 | 0.98-1.01 | 0.505 |
Significant P values are italicized. Open cells indicate confirmed infections; shaded cells indicate suspected infections.
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-25-073
In the rhesus monkeys, sex was not evaluated due to a low number of female monkeys (Table 2). A Poisson regression demonstrated (Table 2) that there was no relationship between the location of the implant and the frequency of confirmed (P = 0.240) or suspected (P = 0.829) catheter infections (but see catheter longevity, below). The number of VAP repairs was significantly related to the frequency of confirmed infections such that a higher number of repairs was related to a higher frequency of confirmed infections (χ2(1) = 27.58, P < 0.001). For every additional repair, the risk of having a confirmed infection increased by 34.5% (95% CI, 20.4%-50.3%) (Table 2, open cells). This relationship was not present (P = 0.734) with the suspected infections (Table 2, shaded cells). Moreover, there was no relationship between confirmed (P = 0.291) or suspected infections (P = 0.555) and how long the catheter was implanted (Table 2). For all models, a Pearson χ2 statistic demonstrated that the data were not overdispersed (χ2/df < 1).
| Independent variable | β | SEM | Odds ratio: Exp(B) | 95% CI | P value |
|---|---|---|---|---|---|
| Sex | NT | NT | NT | NT | NT |
| NT | NT | NT | NT | NT | |
| Location of implant | −0.347 | 0.30 | 0.13 | 0.39-1.28 | 0.240 |
| 0.076 | 0.35 | 1.08 | 0.546-2.15 | 0.829 | |
| Repairs | 0.297 | 0.06 | 1.34 | 1.20-1.50 | <0.001 |
| 0.055 | 0.16 | 1.06 | 0.77-1.45 | 0.734 | |
| Longevity | 0.010 | 0.0099 | 1.01 | 0.99-1.03 | 0.291 |
| −0.011 | 0.02 | 0.99 | 0.95-1.03 | 0.555 |
Significant P value is italicized. Open cells indicate confirmed infections; shaded cells indicate suspected infections.
Abbreviation: NT, Not Tested.
Catheter longevity.
Across 23 years, the average catheter remained patent and implanted in the vein for 22.5 months (SEM = 1.4) in the cynomolgus monkeys (Figure 5A) and 15.5 months (SEM = 1.8) in the rhesus monkeys (Figure 5B). An ANOVA demonstrated that in cynomolgus monkeys there was a significant main effect of catheter implantation site on catheter longevity (F3,295 = 5.39, P < 0.001) such that average longevity was significantly higher when the catheter was implanted in femoral vein when compared with the internal jugular (P = 0.013), external jugular (P < 0.001), and brachial (P = 0.006) veins (Figure 5A). No other significant effects were demonstrated (P > 0.05). Moreover, the sex of the monkey did not significantly influence catheter longevity (P > 0.05). In the rhesus monkeys (Figure 5B), an ANOVA also demonstrated that there was a significant main effect of implant site on catheter longevity (F3,236 = 6.65, P < 0.001). Femoral implants lasted significantly longer than internal jugular implants (P = 0.03) and external jugular implants (P < 0.001) and marginally longer than brachial implants (P = 0.064). No other relationships reached significance (P > 0.05). Finally, note that over the course of the 23 years used to assess catheter longevity and infection rates, no monkey died due to complications related to catheter-tract infections.
Citation: Journal of the American Association for Laboratory Animal Science 2025; 10.30802/AALAS-JAALAS-25-073
Discussion
The main goal of the present study was to determine the longevity and rates of infection of indwelling intravenous catheters and associated VAPs in NHPs self-administering drugs. To accomplish this, 23 years of medical records were examined for female and male cynomolgus and rhesus macaques used in drug self-administration studies at Wake Forest University School of Medicine. The main findings were as follows: (1) the infection rate over 23 years and >500 catheter implants was 13.7% in cynomolgus monkeys and 10% in rhesus monkeys; (2) in cynomolgus monkeys, males were more likely to develop a catheter tract infection compared with female monkeys; (3) the average catheter remained patent and implanted in the vein for 22.5 months in cynomolgus monkeys and 15.5 months in rhesus monkeys; and (4) there was a relationship between the vessel catheterized and catheter longevity, with catheters implanted in the femoral vein lasting significantly longer than those in any other vessel. These findings highlight a significant strength of using Old World macaques, both rhesus and cynomolgus, in long-term, longitudinal studies involving indwelling intravenous catheters.
The present study extended an earlier study by Taylor and Grady15 in which health records from 31 rhesus monkeys (53 catheter implants) were assessed for catheter-tract infections and mean catheter lifespan over an ∼2-year period (January 1, 1996-October 1, 1997). Those monkeys were implanted with indwelling intravenous catheters, but the catheters exited the body in the back, rather than attaching subcutaneously to a VAP as described in the present study. Under those conditions, the investigators reported that 16 of 53 (30.2%) monkeys with catheters were diagnosed with clinical signs of infection and verified by bacterial culture results. If the catheter was uninfected but removed due to concerns about the monkey’s health and/or loss of catheter patency, the mean catheter lifespan was 11.6 months (354 days), which was substantially less than what we report here using the VAP (15.5 months). Of course, over the last 25 years, there have been refinements in surgical techniques, aseptic protocols, and antibiotics (to name a few) that certainly contribute to the better catheter survival rates in the present study compared with this earlier one. However, one of those refinements that we think is critical is the use of VAPs, which resulted in nearly 4 months longer catheter patency than previously reported in individually housed rhesus monkeys in which the catheter exited the body.15
Perhaps most surprising from our analyses was the finding that, in socially housed cynomolgus monkeys, the average catheter remained patent for almost 2 years (22.5 months), which was much longer than observed in rhesus monkeys. An important caveat is that this laboratory began with only rhesus monkeys, so the cynomolgus monkeys had the advantage of refinements in surgical techniques and protocols developed using the rhesus monkeys. Another important consideration is that although catheters were excluded if they were implanted <3 months prior to euthanasia, early research involving the rhesus monkeys was sometimes terminal after several years and, thus, the duration of those catheters was likely underestimated due to the nature of the study. In these cynomolgus monkeys, while both female and male cynomolgus monkeys maintained catheters at similar rates, the males had a significantly higher frequency of infection when compared with females. While there is evidence of sex differences in immune responses (eg, Wilkinson et al24), we think the most likely explanation is that males fought more often than the females, and injuries around the VAP and/or catheter increased the likelihood of infection. Additional research in catheterized, individually housed female and male cynomolgus monkeys needs to be examined to better understand these sex differences in catheter longevity.
Although not systematically assessed in this study, we have noted that antibiotic treatment can slow the progression of confirmed infection, but in cynomolgus monkeys it does not typically prevent the likelihood of eventually removing the catheter due to infection. However, it is an empirical question whether antibiotic treatments would meaningfully extend catheter longevity in instances where infection is suspected but not confirmed. This topic deserves further investigation. Also, note that while maintaining macaques in captivity is expensive, the cost of using the VAP is also high. In particular, specialized needles (Huber needles) are required and these are expensive.25 Over the course of 23 years of using Huber needles and VAPs, we have incorporated techniques to allow for repeated use of a single Huber needle, rather than one-time use, as instructed by the company. After a use in a monkey, the Huber needle is cleaned, placed in an autoclave pouch, and gas sterilized. This process is repeated until the needle appears dull or damaged. As noted from the longevity of our catheter and VAP system (on average nearly 2 years for a single catheter in socially housed cynomolgus monkeys), this technique does not increase the rate of infection.
One important observation from previously published studies is worth reiterating; that is, it is critical that catheter implantation occurs under aseptic conditions. The present data show that once implanted, an indwelling intravenous catheter with a subcutaneous VAP can remain patent for at least 1-2 years. Analysis of >20 years of catheter surgeries identified several risk factors that may contribute to confirmed catheter infections, including the monkey being male and the number of repairs. However, neither the catheter implant location nor the duration of implantation significantly increased the risk of infection. There are several possible reasons why femoral vein catheters lasted longer than catheters placed in other vessels. The vein is large and easy to access, and it is surrounded by muscle that allows for the catheter, secured to the vein, to be covered by muscle prior to skin closure. While much of that is true for jugular veins, one major difference is that all of these monkeys wore aluminum collars, and the physical impact of the collar rubbing on the catheter site may have decreased the longevity of jugular catheters. A possible interpretation of this finding is that the actual rate of infection due to nonsterile handling of the VAP is so low that even performing the procedure hundreds of times over years does not significantly increase the overall infection rate. Furthermore, these findings suggest that minimizing the need for catheter repairs by avoiding puncturing the same point on the skin overlaying the VAP and implementing infection prevention strategies for male monkeys such as leaving them individually housed longer after surgery may help further reduce infection rates. Overall, the results of the present analysis of 23 years of data provide metrics for other investigators using indwelling intravenous catheters for repeated drug use in both NHPs and rodents.
Catheter Implants per Vein. Pie charts depict the total number of catheter implants assessed and a breakdown of the type of catheter implant in cynomolgus (A) and rhesus (B) monkeys.
Relationship Between Catheter Implants and Infection Rates. Pie charts depict the total number of catheter implants assessed and a breakdown of the number of implants that developed a confirmed infection, a suspected infection, or were removed due to blockade/coming out of the vessel in cynomolgus (A) and rhesus (B) monkeys.
Catheter Implants and Infection Rates Across 23 Years of Research. Graphs depict the number of confirmed and suspected infections across 23 y relative to the number of implant surgeries conducted each year and the total number of patent catheters in the laboratory during that year in cynomolgus (A) and rhesus (B) monkeys.
Pie Chart Depicts the Total Number of Confirmed Infections in Cynomolgus Monkeys as a Function of Sex
Catheter Longevity as a Function of Vein. Graphs depict average catheter longevity (months) as well as a breakdown of longevity as a function of implant site in cynomolgus (A) and rhesus (B) monkeys. Each bar represents the mean ± SEM. *P < 0.05; **P < 0.001.
Contributor Notes
Co-first authors.