A Proposal for the Reliable Culture of Borrelia burgdorferi from Patients with Chronic Lyme Disease, Even from Those Previously Aggressively Treated

S. E. Phillips, L. H. Mattman, D. Hulinska, H. Moayad Infection 26 ( 1998) No. 6, pp. 364-367

Summary: Since culture of Borrelia burgdorferi from patients with chronic Lyme disease has been an extraordinarily rare event, clarification of the nature of the illness and proving its etiology as infectious have been difficult. A method for reliably and reproducibly culturing B. burgdorferi from the blood of patients with chronic Lyme disease was therefore sought by making a controlled blood culture trial studying 47 patients with chronic Lyme disease. All had relapsed after long-term oral and intravenous antibiotics. 23 patients with other chronic illness formed the control group. Positive cultures were confirmed by fluorescent antibody immuno-electron microscopy using monoclonal antibody directed against Asp A, and Asp A PCR. 43/47 patients (9l%) cultured positive. 23/23 controls (1OO%) cultured negative. Although persistent infection has been, to date, strongly suggested in chronic Lyme disease by positive PCR and antigen capture, there are major problems with these tests. This new method for culturing B. burgdorferi from patients with chronic Lyme disease certainly defines the nature of the illness and establishes that it is of chronic infectious etiology. This discovery should help to reestablish the gold standard in laboratory diagnosis of Lyme disease.


Lyme disease is a multi-system illness caused by infection with Borrelia burgdorferi. Its manifestations can be myriad. This. coupled with problems in current serologic assays, leads to frequent misdiagnosis at all stages of the illness. Some investigators believe that Lyme Borreliosis is overdiagnosed, while others maintain that it is underdiagnosed. To further confuse matters, a significant percentage of patients with Lyme disease relapse despite antibiotic therapy [1, 2].

Chronic Lyme disease is a controversial topic. Even after extended antibiotic treatment, persistent infection in chronic Lyme disease has been strongly suggested by the persistence of borrelial antigen, as demonstrated by polymerase chain reaction [3, 4]. However, these diagnostic tests are plagued by the absence of a gold standard. The gold standard for laboratory diagnosis in the field of infectious diseases has usually involved culturing the causative organism from the infected host. In the case of Lyme disease, attempts to do so have been disheartening.

The organism has seldom been cultured from cases of treated. Late-stage disease, and if so, primarily from cerebrospinal or synovial fluid [5-8]. Culture of the organism from blood has been a rarity, with successful cultures primarily from cases of untreated, early disease [9, 1O].

We set out to demonstrate a methodology by which we could reliably and reproducibly culture B. burgdorferi from the blood of patients with chronic Lyme disease even though they had had extended antibiotic therapy. If this were successful, it would also provide a unique opportunity to compare the serologic diagnostic criteria set forth by The Centers for Disease Control (CDC) in conjunction with The Association of State and Territorial Public Health Laboratory Directors (ASTPHLD) to what is potentially a gold standard diagnostic test.


The study was a multi-center, controlled blood culture trial with an approximately 2:1 ratio of cases to controls. Patients were selected from private practices in areas both hyper-endemic and non-endemic for Lyme disease. All cases had a diagnosis of Lyme disease and had failed or relapsed after extended oral and intravenous antibiotic therapy. The diagnosis of Lyme disease was made primarily on clinical grounds. Although almost all cases had serologic evidence suggestive of infection with B. burgdorferi, few had positive ELISAs and only a little over half met CDC serologic criteria for Western blot positivity. 4/47 (9%) were positive by Lyme ELISA. 3/47 (6%) were equivocal by ELISA. 26/47 (55%) were positive by CDC criteria for Lyme Western blot. of these, 2O/26 (77% ) were IgM positive, 10/26 (38% ) were IgG positive, and 4/26 (15%) were positive for both IgM and IgG.

To participate in the study, all patients had to have had at least 6 consecutive weeks of therapy with an intravenous third-generation cephalosporin and a subsequent relapse. Some patients had had as long as 6 months of intravenous therapy, with the mean being approximately 3 months. Controls resided in non-endemic areas and consisted of patients with chronic illnesses other than Lyme disease.

The following MPM medium was used for this study: To 1 l of Detroit tap water was added: proteose peptone 2Og, beef infusion from 1,OOOg, dextrose 10g, sodium chloride 10g, dipotassium phosphate 4g, sodium thioglycollate lg, purified agar lg, bacto methylene blue .O04g, sucrose l00g, soluble starch 5g. This was autoclaved for 15 min at 120 degrees C. For the medium to be used in tube or slide culture, it had to be refrigerated for 24 h before final preparation.

For the medium to be used in tubes, l0ml of medium were boiled to dissolve the agar just before use and the following was added to each tube: lml separately autoclaved yeast extract from a 1O% solution to give a final concentration of 1%, and 1 ml of sterile 1O% NaHCO3. Since yeast extract may contain heat-resistant bacilli, it was separately autoclaved for 3O min at 124 degrees C and batch-tested for sterility. The inoculum was 0.lml of blood in EDTA to 4 ml of medium in a slender screw-top tube. Incubation was at 3O degrees C under normal atmospheric conditions for a period of 1-3 weeks.

For the medium to be used in slide culture, it was sterilized in 3Oml amounts in screw-top tubes. Just before use, the medium was boiled to melt the agar and. when cool but not solidified, the following was added: 3ml of separately autoclaved 1O% yeast ex-tract and l0ml of sterile 10% NaHCO3. The broth was then poured aseptically into a sterile plastic Coplin jar. Slides were smeared with the patient's chosen body fluid. The slides had to be specialized so as not to require fixative. The smears were dried in an aseptic environment before being placed in the Coplin jar. once they were inside, the lid was tightly closed and incubation was at 3O degrees C under normal atmospheric conditions for a period of 1-3 weeks.

For the medium to be used for blood agar plates, the broth medium was modified by adding a total of 16 g of agar. Sixty ml of sheep's blood was added as soon as the medium was removed from the autoclave, resulting in "chocolate agar." At this point, separately autoclaved 1O% yeast extract was added to give a final concentration of 1%. The medium was then poured into sterile plastic Petri dishes and stored under refrigeration for 24 h once solidified. The inoculum was O.5ml of blood in EDTA with incubation at 3O degrees C under normal atmospheric conditions for a period of 1-3 weeks. Two blood samples of 5 ml each were collected in EDTA lavender top test tubes from each patient and control. From these, seven cultures were processed from each participant. All positive cultures were stained with acridine orange at pH 3.5 4.O and then confirmed by our laboratory with affinity-adsorbed polyclonal fluorescent antibody to B. burgdorferi (O2-97-91, Kirkegaard & Perry Laboratories, Gaithersburg, MD, USA).

Further confirmation of positive culture results was accomplished by electron microscopy. immuno-electron microscopy utilizing monoclonal antibody directed against Asp A (monoclonal antibody no. 181, courtesy of Prof. B. Left. Stony Brook, NY, USA), and plasmid PCR with Asp A primer. The methods employed in these processes have been previously reported [11, 12].


Of the 47 patients with chronic Lyme disease, 43 (91%) cultured positive for B. burgdorferi. while 23/23 (IOO%) of the controls cultured negative. Many of the cultures were clearly spirochetes when examined under light microscopy (Figures 1-3). Immuno-electron microscopy and Osp A PCR confirmation provided additional confirmatory evidence as to the identity of the spirochetes (Figures 4-7). The slide cultures consistently demonstrated the fastest and most abundant yields. With this technique, placement in the Coplin jar allows for varying gradations of oxygen tension. Sometimes spirochetal growth can be seen after as little as 2O h. appearing as a band near the upper end of the smear.


An attempt to culture B. burgdorferi from the blood of previously aggressively treated chronic Lyme disease patients seemed at first a monumental task. Before undertaking this effort, we therefore had to be as sure as possible that the organisms were indeed present in the blood of these patients.

As a first step, we scrutinized a report where B. burgdorferi had been cultured from the blood of patients with early untreated disease. From this group of patients it had been noted in follow-up that subsequent blood cultures became routinely negative after antibiotic therapy, despite 71% of the patients remaining symptomatic [9]. Three possibilities readily come to mind for the explanation of this paradox: either 1) the infection is cleared, but a post-infectious process continues, or 2) the organism is cleared from the blood rapidly, but finds a pathogenic harbor elsewhere, or 3) the organism is maintained in the blood in an altered state which cannot be cultured on routine media. In response to the first possibility, the notion of a post-Lyme syndrome has countless flaws. A post-infectious syndrome could not explain the observation that patients with "post-Lyme" or "post-Lyme fibromyalgia" responded to re-treatment with antibiotics, only to relapse with its discontinuation [13-15]. With the advent of PCR, antigen capture, and the benefit of those rare successful culture experiments even in the face of prior "curative treatment" [3-8]. the notion of "post-Lyme" should have been dismissed long ago. In response to the second possibility, given the common finding of circulating immune complexes with Lyme disease, we thought this unlikely [16]. Thus we were left with the third and most logical possibility. Specifically, we chose to pursue the organism in its cellwall deficient state, i.e. L-forms, as previously reported [17].

Although L-forms will complex with fluorescent antibody to B. burgdorferi, only as they revert to classic parent forms can the typical spirochetal morphology be seen. There has been a considerable spectrum of cell wall deficiency demonstrated in our laboratory. B. burgdorferi may exist in various forms depending on its environment. In addition to the spirochetal form, we have demonstrated its growth both as amorphous L-forms and rounded giant L-bodies which have been previously described as cystic forms [11, 18]. As B. burgdorferi reverts from cell wall deficiency with the rebuilding of its cell wall, classic spirochetal forms can be seen. Most often, in our cultures, B. burgdorferi can be seen in varying stages of reversion, i.e. some L-dependent spirochetal forms within an L-form colony.

The L-form variants, osmotically fragile by nature, require precise conditions to grow in culture. our medium and methodology are specifically designed for the fostering of cell wall-deficient organisms and their reversion to classic parent forms. In. most instances, the methods must be followed precisely. Even small variations produce no growth. For example, 2% yeast extract instead of 1% is inhibitory, or if the yeast extract is autoclaved with the rest of the medium instead of separately, that too will be inhibitory. However, there is one aspect of B. burgdorferi's growth characteristics which we found to be remarkably non-fastidious. The organism can be easily grown throughout a wide range of pH, from 6.8-7.8. This explains the different ratios of NaHCO3 used in the various types of culture mediums. We are still not sure about the optimal pH for culture. Future research will address that question more specifically.

It should be noted from this study that currently accepted standards for serologic diagnosis seem to be inadequate, only a small minority of participants in the study had positive Lyme ELISAs. Under the current recommendations for two-tier testing by the CDC/ASTPHLD, 91% of the patients in the study would have been misdiagnosed as not having Lyme borreliosis.

It is hoped that our work will help to end a medical controversy which has been going on for far too long. This study proves that chronic Lyme disease is of chronic infectious etiology, and that even antibiotic treatment well in excess of current recommendations is not necessarily curative. Given the flaws in currently accepted serologic diagnostic criteria, it is likely that Lyme borreliosis is vastly underdiagnosed. May this research help to focus the scientific community on effective curative therapies for patients with chronic Lyme disease.

It should also be noted that, in addition to its utility in growing B. burgdorferi, the MPM medium may be useful for culturing a variety of other spirochetes from patients. So far, we have clutured several types of spirochetes from over 100 patient cases. Reports on these series will be forthcoming.


1 Krupp. L. B.. Maser, V.. Schwartz, J., Doyle, P. K., Langenback, L. J., Fernquist, S. R.: Cognitive functioning in late Lyme borreliosis Arch. Neurol. 48 (1991) 1125-1129

2 Logigian. E. L.. Kaplan. R. F., Steere, A. C.: Chronic necrologic manifestations of Lyme disease N. Engl J. Med. 323 (1990) 143~1444.

3 Bayer. M. E.. Zhang, L.. Bayer, M. H.: Borrelia burgdorferi DNA in the urine of treated patients with chronic Lyme disease symptoms A PCR study of 97 cases Infection 24 (1996) 347-353

4 Nocton, J. J.. Dressier. F.. Rutledge, B. J., Rys, P. N., Persing, D. H., Steere, A. C.: Detection of Borrelia burgdorferi DNA by polymerase chain reaction in svnovial fluid from patients with Lyme arthritis N. Engl. l Med. 330 (1994) 229-234

5 Preac Mursic, V., Weber, K., Pfister, H. W., Wilske, B.. Gross, B., Baumann, A., Prokop, J.: Survival of Borrelia burgdorferi in anti-biotically treated patients with Lyme borreliosis Infection 17 (1989) 355-3S9

6 Schmidli, J., Hunzicker, T., Moesli, P., Schasd, U. B.: Cultivation of Borrelia burgdorferi from joint fluid three months after treatment of facial palsy due to Lyme borreliosis I Infect Dis 158 (1988) 905-906

7 Pfister, H. W.. Preac Mursic. V.. Wilske, B.. Schielke, E.. Sorgel, F., Einhaupl, K. M.: Randomized comparison of ceftriaxone and cefo-taxime in Lyme neuroborreliosis. J. Infect. Dis 163 (1991) 311-318.

8 Hassler, D.. Riedel. K., Zorn, J., Preac Mursic, V.: Pulsed high-dose cefotaxime therapy in refractory Lyme borreliosis (letter). Lancet 338(1991) 193

9 Nadelman, R. B.. Pavia, C. S., Magnarelle, L. A., Wormser, G. P.: Isolation of Borreha burgdorferi from the blood of seven patients with Lyme disease. Am. J. Med. 88 (199O) 21-26.

10 Berger, B. W., Johnson. R. C., Kodner, C., Coleman, L.: Cultivation of Borrelia burgdorferi from the blood of two patients with erythe-ma migrans lesions lacking extracutaneous signs and symptoms of Lyme disease. l. Am. Acad Dermatoi. 3O (1994) 48-51

11. Hulinska, D., Bartak, P., Hercogova, J., Hancil, J., Basta, l., Schram-lova, J.: Electron microscopy of Langerhans cells and Borrelia burg-dorferi in Lyme disease patients. Zbl. Bakt 28O (1994) 348-359

12 Hulinska, D., Krausova, M., Janovska, D., Rohacova, H.. Hancil, J., Mailer, H.: Electron microscopy and the polymerase chain reaction of spirochetes from the blood of patients with Lyme disease. Cent Eur. J. Public Health I (1993) 81-85

13 Sigal, L. H., Patella. S. J.: Lyme arthritis as the incorrect diagnosis in pediatric and adolescent fibromyalgia Pediatrics 9O (1992) 523-528

14 Dinerman, H., Steere, A. C.: Lyme disease associated with fibromy-algia. Ann Intern Med. 117 (1992) 281-285

15 Steere, A. C., Taylor, E., McHugh, G. L., Logigian, E. L.: The over-diagnosis of Lyme disease JAMA 269 (1993) 1812-1816

16. Schutzer, S. E., Coyle, P. K., Belman, A. L.. Golightly, M. G., Drul-le, J: Sequestration of antibody to Borrelia burgdorferi in immune complexes in seronegative Lyme disease Lancet 33S (1990) 312-315

17 Preac Mursis V.. Wanner, G., Reinhardt, S., Wilske. B.. Busch, U., Marget, W.: Formation and cultivation of Borrelia burgdorferi spheroplast L-form variants Infection 24 (1996) 218-226

18 Brorson, 0, Brorson, S. H.: Translormation of cystic forms of Borrelia burgdlJreri to normal. mobile spirochetes Infection 2S (1997) 240-246

Received: 27 September 1997/Revision accepted: 3 September 1998

S. E Phillips, M. D., Greenwich Hospital, S Perry Ridge Rd., Greenwich, CT O683O

L. H. Mattman, Ph. D., Spirotech Institute, Empire State Bldg., 350 Fifth Ave.. Suite 61OI, New York, NY 10118

H. Moayad, D. O., Columbia North Hills Medical Center, 4401 Booth Calloway Rd., North Richland Hills, TX 76180

Dagmar Hulinska, Ph. D., National Institute of Public Health. GEM-ELM. Srobarova 48. l0042 Praha IO. Czech Republic.

Correspondence to: Dr. S. E. Phillips. 10 Roberts Lane. Suite 2. Ridgefield. CT 06877. USA.