Femoral Vein Cannulation in the Treatment of Osteomyelitis

Objective. The main objective was to investigate the feasibility of using the femoral vein for long-term venous access. This was accomplished in the course of treating osteomyelitis patients, using a combination of long-term (6 weeks), outpatient, intravenous (IV) antibiotics administered through a femoral central line. This was combined subsequently with the use of hyperbaric oxygen (HBO) therapy. Using the femoral vein for central venous access means there is no risk of creating an iatrogenic pneumothorax, which would prohibit the subsequent use of HBO therapy. In addition, the propitious nature of the groin skin flora, Propionibacteriaceae, seems less inclined to participate in biofilm production, the root cause of central line infections. Methods. The femoral central lines were all inserted in the operating room (OR) and handled like a regular outpatient surgery with a meticulous skin prep completed by experienced OR nurses. Experienced technicians assisted with the surgery. After insertion, the lines were then specially secured to prevent sliding. Vancomycin was administered preoperatively. Results. Surprisingly, femoral lines placed in this way remained free of infection for up to 70 days. Eight patients with Wagner stage 2 ulcers and underlying osteomyelitis were treated with a course of 6 weeks of vancomycin, followed by HBO therapy. A cure rate of 75% was achieved. Conclusion. Femoral vein cannulation for antibiotic administration is safe and effective in treating patients with osteomyelitis. The lines should be placed in the OR with the help of skilled personnel. Femoral cannulation avoids the risk of pneumothorax, and the lines can be used for up to 70 days. By avoiding pneumothorax, the patients remain candidates for HBO therapy.

The use of long-term, outpatient, antibiotic administration to treat osteomyelitis is an important part of wound care. This article reports the surprising finding that femoral venous access is a safe and efficacious method of delivering antibiotics for relatively long periods of time (up to 70 days).

A small cohort of 8 patients with diabetes is described. They initially presented to the authors’ Wound Care Center with Wagner stage 2 ulcers and were subsequently found to have underlying osteomyelitis. They were chosen for this study because of their need for a preliminary, 6-week course of intravenous (IV) antibiotics and their subsequent need for adjunctive hyperbaric oxygen (HBO) treatment. Hyperbaric oxygen therapy was used, because there had not been a distinct improvement in the patients’ scans after 4 weeks of antibiotic therapy.¹ A history of a pneumothorax is a contraindication to hyperbaric medicine² and unfortunately iatrogenic pneumothorax complicates more than 6% of subclavian vein central line placements.³ Furthermore, it is the authors’ opinion that this issue probably occurs more commonly than even this, but does not appear in the literature because of the reluctance of physicians to report untoward incidents. The creation of a pneumothorax defeats the object of the exercise, as it precludes the therapeutic combination of HBO therapy and antibiotics. More so, the onset of a pneumothorax may be delayed several days⁴ and go undetected in the immediate postoperative chest X-ray.  

To avoid the risk of a pneumothorax, the authors started using the femoral vein as an access to the central venous system. Indeed, the main focus of this paper is to document the surprising finding that femoral lines inserted in the operating room (OR) with careful attention to technique can be safely left in place for long periods of time with minimal complications. The authors were also encouraged by a recent study⁵ that showed no difference in infection rates between subclavian and femoral lines.

In addition, Marnejon et al⁶ found that among the risk factors for upper extremity venous thrombosis associated with peripherally inserted central venous catheters (PICC lines) was the infusion of antibiotics, specifically vancomycin. Since vancomycin is the lynchpin of this study’s therapy, this finding further dissuaded the authors from using PICC lines. 

Furthermore, not widely appreciated by the surgical community is the impression that the main cause of line sepsis is biofilm production.⁷ Fortunately, the bacterial flora of the skin of the groin, predominantly Propionibacteriaceae, lacks the propensity for participation in biofilm production⁸ and, ipso facto, are not virulent.   

Detailed description of technique for femoral central line insertion. Since the avoidance of line sepsis is critically important, all procedures were completed in the OR, and they were treated like any other outpatient surgical procedure, with monitored anesthesia care, the administration of 1 g vancomycin preoperatively, and meticulous skin preparation. This consisted of clipping all hair from the areas over the femoral veins, then 8 minutes of scrubbing with betadine (Wet Premium Skin Scrub Trays, Medline, Mundelein, IL) from umbilicus to knees, and finally a 30-second final scrub in the groin creases with ChloraPrep Skin preparation (CareFusion, San Diego, CA)  

All surgeries were performed by the first author at Sandhills Regional Medical Center, Hamlet, NC, with the skilled assistance of an experienced OR technician. A Siemens Acuson ultrasound machine (Siemens, Washington, DC), with a 6 L 3 probe in a sterile sheath, was used to confirm common femoral vein patency and to aid in needle insertion. The 3-mL syringe and the 2.5-inch XTW needle from the ARROWg+ard Blue PLUS kit (Arrow International, Reading, PA) were used to access the vein after infiltration of local anesthetic. At this point, the Spring-Wire Guide from the same kit (Arrow International, Reading, PA) was advanced to the right atrium to confirm its location in the inferior vena cava. A Phillips BV Pulsera SN 002286 C-arm (Phillips Healthcare, Andover, MA) was used to follow the progress of the wire and to check for any unusual anatomy. In about 50% of cases, this wire would not advance easily, so a 0.025-in ZIPwire (Boston Scientific, Marlborough, MA) was used instead. 

Then, after a small skin incision had been made, a 7F Super Sheath (Boston Scientific, Marlborough, MA) was advanced over the wire and withdrawn to make a tunnel to the vein. This was followed expeditiously by the ARROWg+ard Blue PLUS Catheter. After which an inferior vena cava venogram was recorded to confirm normal vena cava anatomy, and that the position of the catheter tip in the center of the lumen of the common iliac vein, or inferior vena cava, was correct with rapid, unimpeded flow of the dye.

The most critical part of the procedure was the securing of the catheter to prevent sliding of the tubing, and the associated risk of dragging bacteria under the skin and into the tunnel, leading to the vein. This was done by making a purse string around the skin access site using 2-0 nylon (Figure 1). This was tightened to about 5 lb tension, not too tight, to avoid creating a rim of ischemic tissue, but tight enough to prevent the catheter from sliding. Then the same suture was looped around the distal part of the catheter hub and tightened to the point where the catheter was held snugly against the skin and could not move (Figure 2). Two other sutures were placed using the holes provided by the manufacturer; again, the direction of the sutures was such that they also helped prevent sliding (Figure 3).  

Vancomycin administration. Patients then returned the following day and were scheduled to come daily thereafter for 40 more days for vancomycin administration in the outpatient department. Each day the IV site was inspected by experienced nurses and cleaned with betadine (Purdue Pharma Products, Stamford, CT) and ethyl alcohol. On Mondays and Fridays, the entire dressing was changed and a fresh Hypafix Dressing Retention Tape (Smith & Nephew, St. Petersburg, FL) was applied. The pharmacy department monitored the antibiotic levels and adjusted the dose to maintain the optimal serum concentration.  

Bone scan and osteomyelitis diagnosis. Three-phase bone scans were performed using the Alegent Health Midlands Hospital imaging protocol (Papillion, NE)⁹ and were read independently by the hospital radiologist. In 2 cases, the initial diagnosis of osteomyelitis was made from a bone biopsy; the other 6 cases had positive bone scans. 

Osteomyelitis was judged to be cured when the 3-phase bone scan was negative (4 patients). According to Nikpoor,¹⁰ “A negative bone scan essentially rules out osteomyelitis; there is no need to proceed with a WBC [white blood cell] scan if the bone scan is normal.” In the absence of a scan (Patient GJ refused to have one), if there has been no clinical sign of recurrence for more than a year, the authors judged the patient was cured. 

Patient MF transferred to another wound care center after his ulcer was cured; he relayed to the authors that he has had no recurrence and is presumed cured of osteomyelitis.  

Hyperbaric medical oxygen therapy. The HBO therapy protocol involved 2 atmospheres absolute (2 ATA) for 90 continuous minutes, in a Perry Sigma Monoplace Hyperbaric Cylinder, (34-100-PVHO=BP-S3400; Perry BaroMedical, Riviera Beach, FL). The patients were treated Monday through Friday for 8 weeks, a total of 40 dives.

Patient population. All patients had type 2 diabetes with Wagner stage 2 lesions, most often involving the toes. Patients’ ranged in age from 50 years to 71 years; there were 5 males and 3 females. All patients were treated with 40 dives each at the Sandhills Center for Wound Healing and Hyperbaric Medicine. All signed informed consent to use their data for research purposes.  

At the completion of HBO therapy, a series of silver impregnated dressings, gradually changing over to collagen dressings, was used to treat the skin ulcers until they closed.

Surface infections were treated with the appropriate oral antibiotic, and oral sulfa and ciprofloxacin were continued until osteomyelitis was cured.  

Femoral lines were used for up to 70 days, (range 38-70 days). No lines became infected. A complication that occurred was mechanical: in 1 case (patient GJ) the line was inadvertently pulled out and was replaced in the other femoral vein (Table 1).  

Patient SB did not want IV antibiotics initially, so there was a delay of more than 3 months while he took oral antibiotics.  

In 2 cases, there was an inordinate delay between the first vancomycin infusion and the start of HBO treatment. In the case of GJ, the delay was caused by the need to address a heart condition that developed; a subsequent car wreck compounded the problem.  

Patient BS needed angioplasty revascularization of his left leg prior to HBO treatment, then was admitted to the hospital with a change of mental status, and finally the combination of his noncompliance with his family’s unwillingness to arrange transportation added months to the interval between his first vancomycin treatment (Feb. 5, 2014) and the start of HBO therapy (Sept. 2, 2014).  

Patient DV had been receiving oral antibiotic treatment for months from another physician prior to her referral to the Wound Care Center. A bone culture came back positive November 11, 2014, so she was able to start her HBO treatment on December 9, 2014 before receiving her femoral line January 7, 2015.  

In 3 cases, patient MF, MM, and DV, the lines were left in significantly longer than usual, primarily because of missed infusions and the need to make up for these. Lack of compliance, weather conditions, and lack of transportation were the main reasons for missed appointments. Note: The authors were surprised at the dedication of most of these patients with refractory osteomyelitis who require daily attendance for so many months, and who have to travel long distances to get to the clinic, despite the high price of gasoline and the poverty of this patient population. The authors understood why patient KL simply gave up and refused further treatment other than palliation. Similarly, patient BS was admitted to hospice care and currently also receives palliative care.  

After his skin ulcer healed, patient MF transferred to a wound care center closer to his home; when the authors’ secretary contacted the clinic, they informed her he was cured of his osteomyelitis, but could not give an exact date.

From the time of ulcer presentation to healing, it took between 8 and 30 weeks and between 20 and 38 weeks for resolution of the osteomyelitis (Table 2). In 2 cases, patients KL and BS, the ulcer did not heal, nor did the osteomyelitis. In 4 cases, there was radiological proof the osteomyelitis was cured; in the cases of GJ and MF a cure was assumed, because there had been no recurrence of clinical signs or symptoms for more than a year.  

This results in a 75% cure rate for the osteomyelitis component, and a 75% cure rate for the ulcer. Patient DV subsequently developed osteomyelitis in another area and is currently undergoing treatment for this second problem.  

Bloodstream infections from central lines can best be described as a medical disaster, occurring in 250 000 patients in the United States each year and causing a staggering 62 000 deaths.¹¹ A single incident of such an infection can cost $56 000 to treat.¹¹

For these reasons, the authors believe central line insertion should be taken seriously, and the procedure should be performed in the OR with the help of personnel experienced in maintaining a sterile field, not at the bedside. 

In connection with groin cannulation, a concern is the proximity of the perineum to the groin would result in increased bacterial cross-contamination. However, a study of the skin flora of this area shows bacteria in the groin are a stable population and quite different from perineal flora.¹² 

The authors’ personal observations are that the route of perineal drainage shows that in the recumbent patient the flow is from the groin downward toward the perineum; and when the patient stands, the drainage is down the inside of the thigh away from the groin. By way of contrast, with the patient’s head on a pillow, the subclavian insertion site is in direct line of oral or nasal secretions from coughing or sneezing, especially when the recumbent patient is resting on his side. In addition, the negative intrathoracic pressure engendered by respiration would tend to draw any such surface-infected fluids inward, alongside the catheter. Nasal and oral bacteria are particularly prone to biofilm production. 

Another safety consideration that makes the femoral site attractive is that if the catheter cannot be advanced up the right side during surgery, then it is safe to proceed to try the left side during the same anesthesia. Conversely, the risk of creating bilateral pneumothoraces should preclude this dual approach by a prudent surgeon working in the subclavian area, thus causing delay and increased expense.  

The femoral approach is also easier for the surgeon in an ever-increasing population of obese patients, as the depth of the femoral vein in the thigh stays relatively constant.¹³ On the other hand, rampant obesity increasingly obscures landmarks in the thoracic area. From the jaw line downward, the morphology of some patients increasingly approaches that of a truncated cone of adipose tissue, and surgery becomes correspondingly more difficult. Should a chest tube be required, this same thickness of fat overlying the rib cage can also make this procedure difficult.

In a small hospital, a pneumothorax complication negatively impacts the program’s reputation, and leaves it open to legal action; and because it was “hospital acquired,” the costs are not reimbursed. Transferring a patient for a thoracoscopic surgery to repair a leaking lung may cost upward of $20 000 dollars.¹⁴ Also, for these reasons the use of the femoral approach should gain increasing acceptance.  

A recent article on PICC lines,¹⁵ specifically used to treat osteomyelitis,reported a mean duration of catheterization of 21 days, but an 8% infection complication rate. However, it should be noted that the use of PICC lines is becoming increasingly popular; about 3 million are placed annually, 70% by nurses and usually at the bedside,¹⁶ often in cancer patients, and for total parenteral nutrition. In addition, claims have been made that such lines can be kept in place for 1 year.¹⁷ 

Johannsen et al¹⁸ published a systematic review of the literature and concluded: “scientific evidence supporting any advantage or disadvantage of PICC when comparing PICC with traditional central venous lines is limited, apart from a tendency towards increased risk for DVT…with PICC.” In this regard, to date the authors have not experienced venous thrombosis as a complication of femoral venous cannulation. 

In this small series of patients, the time from clinic presentation to ulcer closure was between 8 and 30 weeks; and the time from diagnosis to cure of osteomyelitis was between 23 and 38 weeks. The percent cured using an initial combination of outpatient IV antibiotics, and HBO therapy was 75%. This falls in the 60% to 85% national range for cure by this combined technique, as outlined in a review by Hart.¹⁹ 

The most important finding of this research concluded that if inserted in the OR, femoral lines are safe for long-term antibiotic administration and do not run the risk of expensive complications or of interfering with HBO treatment by causing iatrogenic pneumothorax. Of all branches of surgery, wound care specialists are most aware of biofilm dangers; so this may translate into a greater use of femoral access where the skin flora seems less conducive to infection.  

From the Sandhills Center for Wound Healing and Hyperbaric Medicine; and Health Information Department, Sandhills Regional Medical Center, Hamlet, NC 

Address correspondence to:
Alan S. Coulson, MD
Hamlet PPM, LLC
Sandhills Surgical
108 Endo Lane, Suite 3
Hamlet, NC 28345
alan.coulson@sandhillsregional.com

Disclosure: The authors disclose no financial or other conflicts of interest.