Article Text

PDF

Serum procalcitonin measurement for detection of intercurrent infection in febrile patients with SLE
  1. K C SHIN,
  2. Y J LEE,
  3. S W KANG,
  4. H J BAEK,
  5. E B LEE,
  6. H A KIM,
  7. Y W SONG
  1. Department of Internal Medicine
  2. Clinical Reseach Institute
  3. Seoul National University College of Medicine
  4. Seoul, Korea
  1. Dr Y W Song, Department of Internal Medicine, Seoul National University College of Medicine, Chongro-ku, Yongon-dong 28, Seoul 110–744, Korea ysong{at}plaza.snu.ac.kr

Statistics from Altmetric.com

It is sometimes difficult to distinguish infection from disease flare in febrile patients with systemic lupus erythematosus (SLE). Chill, leucocytosis, and increased C reactive protein (CRP) are known to be markers favouring infection.1 Procalcitonin (PCT) is the precursor of calcitonin and is synthesised in the parafollicular C cells of the thyroid. Serum PCT increases in severe bacterial or fungal infection but does not increase, or increases only slightly, in viral infections.2 3 The purpose of this study was to evaluate the usefulness of serum PCT in febrile episodes of patients with SLE to distinguish infection from disease flare.

We prospectively enrolled 19 patients with SLE with fever who were admitted to Seoul National University Hospital between October 1998 and April 1999. Fever was defined as an axillary temperature over 38°C. Eleven patients with inactive SLE were enrolled as controls. Blood of the febrile lupus patients was withdrawn three times: on the day of the hospital visit, and after 24 hours and 48 hours. Another sample was withdrawn two weeks after defervescence to control infection or because of a decrease in lupus activity. At the detection of fever, blood cultures and other necessary cultures were performed with complete blood count, Westergren erythrocyte sedimentation rate (ESR), CRP, serum anti-dsDNA, complements (C3, C4), urine analysis, serum creatinine, and chest x ray examination.

The patients were divided into groups on the basis of viral infection, non-viral infection, and lupus flare. Lupus flare was defined by the Systemic Lupus Erythematosus Disease Activity Index (SLEDAI)4 as an increase of more than three points compared with the SLEDAI of the patient one month before the febrile period. Serum PCT was measured by an immunoluminometric assay (LUMItest, Brahms Diagnostika, Berlin). Twelve lupus patients were shown to have infection (nine non-viral, three viral infections) and seven patients had lupus flare. Non-viral infections consisted of urinary tract infection (n=3), sinusitis (n=1), tuberculosis (n=1), scrub typhus (n=1), pseudomembranous colitis (n=1), aspergillosis (n=1), and nocardiosis (n=1). Viral infections were upper respiratory infection (n=2) and gastroenteritis (n=1).

The white cell count was higher in the group with non-viral infection than in the group with lupus flare (p=0.015, table 1). Westergren ESR increased in 89% of the groups with viral and non-viral infection and in all members of the group with lupus flare. There was no difference between the Westergren ESRs of the groups with viral and non-viral infection and lupus flare during the early febrile period (p=0.32). CRP increased in 92% of the group with infection and 89% of the group with lupus flare. CRP tended to be higher in the group with non-viral infection than in the groups with viral infection or lupus flare, but this did not reach significance (p=0.98).

Table 1

Baseline data of 19 patients with systemic lupus erythematosus (SLE) during the early febrile period. Results are shown as means (SD)

Serum PCT in the group with non-viral infection tended to increase continuously or rise gradually in the early febrile period (fig 1). Serum PCT levels in the early febrile period were significantly higher in the group with non-viral infection (mean (SD), 0.98 (0.12) ng/ml) than in the group with viral infection (0.13 (0.04) ng/ml, p<0.01), the group with lupus flare (0.24 (0.18) ng/ml, p<0.01), and the controls (0.12 (0.03) ng/ml, p<0.01). There was no statistical difference in the PCT levels of the groups with viral infection or lupus flare and the control group (p=0.12). After defervescence the serum PCT level was 0.21 (0.18), 0.13 (0.04), 0.14 (0.08) ng/ml in the group with non-viral infection, the group with viral infection, and the group with lupus flare, respectively. None of the values were significantly different from the control group (p=0.86, 0.49, 0.58, respectively).

Figure 1

Mean procalcitonin levels during and after febrile episode. Bars denote standard error.

New serological markers to distinguish lupus flare from infection have been investigated in an attempt to differentiate infection with lupus flare. Serum PCT is normally almost undetectable (<0.1 ng/ml) and not influenced by kidney function. Dandona et alfound that after injecting endotoxin intravenously, PCT increased suddenly at six hours and maintained a plateau for more than 24 hours.5 Recently, the stimulation of peripheral blood mononuclear cells with lipopolysaccharide was found to increase PCT mRNA transcription.6 PCT levels in autoimmune disease were first studied by Eberhard et al, who showed that the PCT increased in vasculitic patients with infection (1.93 (1.19) ng/ml) and decreased after control of the infection (0.63 (0.62) ng/ml).7 PCT did not increase in patients with Wegener's granulomatosis upon disease aggravation but increased with combined infection.8

Our study is the first to observe serum PCT changes in lupus patients with fever and defervescence prospectively. In our study lupus patients with bacterial or fungal infection had higher serum PCT levels than those with viral infections and a higher level than the controls. We tried to determine the serum PCT changes during the febrile period by measuring serial samples. Serum PCT in the group with non-viral infection tended to increase continuously or rise gradually in the early febrile period (fig 1). The PCT values varied among the non-viral group and serious infections, such as aspergillus pneumonia, showed higher values than urinary infection or sinusitis (data not shown). The pitfalls of PCT as a marker for infection are that it may not increase or increase only slightly in viral infection. Our study showed that there was no difference between the serum PCT of the group with viral infection and the control group.

CRP is useful for detecting and differentiating infections in lupus.9 CRP rises earlier and is more sensitive than ESR. In this study, CRP tended to increase in the case of non-viral infection, compared with viral infection or lupus flare, but this did not reach statistical significance. Our results indicate that during the early febrile period, serum PCT increased significantly in patients with SLE with non-viral infection compared with patients with lupus flare. Serum PCT decreased after defervescence. These results suggest that serum PCT helps in detecting bacterial or fungal infections during the early febrile period in SLE.

Acknowledgments

We are indebted to Bukyung Co and Miss Kyung Hee Lee for their help during this study.

References

View Abstract

Footnotes

  • This study was supported by a grant from Seoul National University, Clinical Research Institute, Institute of Allergy and Clinical Immunology.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.