- Email: [email protected]
Clinical factors indicating the presence of pyogenic liver abscess in community-acquired Klebsiella pneumoniae bacteremia
369
Letters to the Editor nosis) was demonstrated in 63 (38.4%) patients, followed by sialoadenitis in 16 (9.8%) and mucoepidermoid carcinoma in nine (5.5%) patients. MSG tuberculosis was diagnosed in five (3%) patients by either histopathologic study or mycobacterium tuberculosis PCR. Their mean age was 37.4 (range 25—49) years. All patients were female and midwives. The involved MSG was the submandibular gland in four (80%) patients and the parotid gland in only one (20%) patient. None of the patients had a past or family history of TB. Physical examination was unremarkable except for the presence of MSG masses. All patients received standard anti-tuberculous chemotherapy for 8—12 months. Iran is an endemic country for TB, so it is important to have an estimate of TB prevalence in MSG masses. To the best of our knowledge, this is the first formal report in this regard. A recent study in South Korea reported eight cases of MSG tuberculosis among patients admitted at any of the three tertiary referral centers in a period of 10 years.11 The parotid and submandibular glands were involved in five (62.5%) and three cases (37.5%), respectively. Another study found only one case of salivary gland TB among 150 patients with salivary gland disease in a 14-year experience in India.12 TB involvement was located in a minor salivary gland in that study. In another study of 155 salivary gland masses, 10 (6.4%) cases of TB infection were found.13 A study in Turkey reported parotid involvement of TB in six cases among 216 patients with a parotid mass.14 Finally, in a 12-year study in China, nine patients were reported to have suffered from tuberculosis of the parotid glands.15 The prevalence of MSG tuberculosis in our study is relatively similar to that in other reports from Asia. Considering the extremely rare event of tuberculous involvement of the submandibular gland,3,4 our experience is unusual in that 80% of our patients had their submandibular glands involved. We doubt that this unusually high rate of submandibular gland tuberculosis could be regarded as incidental. However, we are at present unclear as to the mechanism of TB’s strong tendency to involve the submandibular glands in our patients. Studies with larger sample sizes are needed to clarify the significance of this finding. Conflict of interest: No conflict of interest to declare.
References 1. Choudhury N, Bruch G, Kothari P, Rao G, Simo R. 4 years’ experience of head and neck tuberculosis in a south London hospital. J R Soc Med 2005;98:267—9. 2. Kundu S, Das S, Dey A, Sengupta A. Tuberculosis of parotid gland–— a rare clinical entity. Indian J Otolaryngol Head Neck Surg 2004;56:57—8.
Clinical factors indicating the presence of pyogenic liver abscess in community-acquired Klebsiella pneumoniae bacteremia Klebsiella pneumoniae is a leading cause of pyogenic liver abscess (PLA) in some Asian countries,1—5 where one of the most common sources of K. pneumoniae bacteremia is PLA.6—8
3. Manipoud P, Rerolle S, Kemeny JL, Lafaye M. Tuberculosis of the submandibular gland. Apropos of a case and review of the literature. Ann Otolaryngol Chir Cervicofac 1993;110:482—4. 4. Bodner L, Lewin-Epstein J, Shteyer A. Submandibular tuberculous lymphadenitis (scrofula): report of two cases. J Oral Maxillofac Surg 1990;48:192—6. 5. Mignogna MD, Muzio LL, Favia G, Ruoppo E, Sammartino G, Zarrelli C, et al. Oral tuberculosis: a clinical evaluation of 42 cases. Oral Dis 2000;6:25—30. 6. Rowe-Jones JM, Vowles R, Lighton SE, Freedman AR. Diffuse tuberculous parotitis. J Laryngol Otol 1992;106:1094—5. 7. Tuli AS, Gupta V, Singh H, Chary G, Chand AK. Primary tuberculosis of parotid gland. Indian J Otolaryngol Head Neck Surg 2005;57:82—3. 8. Suleiman AM. Tuberculous parotitis: report of 3 cases. Br J Oral Maxillofac Surg 2001;39:320—3. 9. Lee IK, Liu JW. Tuberculous parotitis: case report and literature review. Ann Otol Rhinol Laryngol 2005;114:547—51. 10. Hamdan AL, Hadi U, Shabb N. Tuberculous parotitis: a forgotten entity. Otolaryngol Head Neck Surg 2002;126:581—2. 11. Kim YH, Jeong WJ, Jung KY, Sung MW, Kim KH, Kim CS. Diagnosis of major salivary gland tuberculosis: experience of eight cases and review of the literature. Acta Otolaryngol 2005;125:1318—22. 12. Tewari M, Shukla HS, Kumar M, Sharma OP. Non-neoplastic salivary gland disease with reference to minor salivary gland tuberculosis. Indian J Surg 2003;65:168—71. 13. Shaha AR, Webber C, DiMaio T, Jaffe BM. Needle aspiration biopsy in salivary gland lesions. Am J Surg 1990;160:373—6. 14. Suoglu Y, Erdamar B, Colhan I, Katircioglu OS, Cevikbas U. Tuberculosis of the parotid gland. J Laryngol Otol 1998;112: 588—91. 15. Chou YH, Tiu CM, Liu CY, Hong TM, Lin CZ, Chiou HJ, et al. Tuberculosis of the parotid gland. J Ultrasound Med 2004;23: 1275—81.
Pantea Hashemi Armin Rashidi* Ebrahim Razmpa Department of Otolaryngology, Imam Khomeini Hospital, Tehran University of Medical Sciences, Tehran, Iran *Corresponding author. Address: No 12, Aramesh Alley, Mellat Park, Tehran, Iran. Tel.: +98 21 22042595 E-mail address: [email protected] (A. Rashidi) Corresponding Editor: Michael Whitby, Brisbane, Australia 17 July 2006 doi:10.1016/j.ijid.2006.09.006
PLA presents a diagnostic challenge, because half of all patients show no specific symptoms or signs that direct attention to the liver. This study was conducted to identify clinical and laboratory factors indicating the early diagnosis of PLA in community-acquired K. pneumoniae bacteremia. Patients at three hospitals in Korea, diagnosed with K. pneumoniae bacteremia between January 2001 and December 2004, were retrospectively reviewed with regard to
370 Table 1
Letters to the Editor Clinical characteristics of patients with community-acquired Klebsiella pneumoniae bacteremiaa
Variables
PLA group (N = 38)
Non-PLA group (N = 93)
p Value
Age (mean standard deviation) years Male:female
62.1 12.9 1.2
62.2 13.8 1.3
1.0 0.9
Underlying conditions Pancreaticobiliary disease including tumor Liver cirrhosis with/without hepatoma Diabetes mellitus None
7 (18.4%) 0 9 (23.7%) 25 (65.8%)
27 12 25 41
0.2 0.02 0.7 0.04
Symptoms Right upper quadrant pain
9 (23.7%)
28 (30.1%)
0.5
Physical findings Body temperature >38 or <36 8C Hepatomegaly
24 (63.2%) 1 (2.6%)
54 (58.1%) 4 (4.3%)
0.6 1.0
Laboratory findings White cell count >12 or <4 109/l Platelet <100 109/l C-reactive protein >10 mg/dl Alanine transaminase >100 U/L Albumin <3 g/dl Total bilirubin >1.5 mg/dl Alkaline phosphatase >300 U/L Lactate dehydrogenase >400 U/L Chest X-ray abnormalities b
19 (50.0%) 16 (42.1%) 28/35 (80.0%) 17 (44.7%) 15/37 (40.5%) 14 (36.8%) 3/37 (8.1%) 22/27 (81.5%) 6 (15.8%)
45 (48.4%) 33 (35.5%) 49/74 (66.2%) 34 (36.6%) 39/90 (43.3%) 58 (62.4%) 8/89 (9.0%) 32/61 (52.5%) 13 (14.0%)
0.9 0.5 0.1 0.4 0.8 0.008 1.0 0.01 0.8
Severity of illness SAPS II score of 30 Septic shock
8 (21.1%) 5 (13.2%)
24 (25.8%) 12 (12.9%)
0.57 1.0
(29.0%) (12.9%) (26.9%) (44.1%)
SAPS, simplified acute physiology score. a Not all parameters are presented. b Chest X-ray abnormalities in the right lower lung field included pleural effusion, atelectasis, or elevated right hemidiaphragm.
demographic, clinical, and laboratory data. Healthcare-associated and nosocomial cases were excluded because K. pneumoniae bacteremia associated PLA is usually community-acquired.1 Only community-acquired cases in adults who had undergone abdominal ultrasonography (US) or computed tomography (CT) were included. The severity of acute illness was assessed at the time of initial bacteremia by using the simplified acute physiology score (SAPS) II.9 Among a total of 311 adult patients with K. pneumoniae bacteremia, 131 met the inclusion criteria and were analyzed. In 38 of the 131 patients, K. pneumoniae bacteremia was associated with PLA (PLA group). PLA was cryptogenic in 31 patients and of biliary origin in seven. In 93 patients (non-PLA group), the sources of bacteremia were biliary tract (49), unknown source (17), urinary tract (16), lung (5), peritoneum (3), bone and joint (2), and skin and soft tissue (1). Bacteremia was polymicrobial in five patients only, in the non-PLA group ( p = 0.32). Clinical characteristics are compared in Table 1. Multiple logistic regression revealed that only hyperbilirubinemia (total bilirubin >1.5 mg/dl) was negatively associated with PLA (odds ratio 0.16; 95% confidence interval 0.04—0.73; p = 0.02). Diabetes mellitus and pancreaticobiliary disease were the most common underlying conditions in our patients, but no significant difference was found between the PLA group and nonPLA group. Several symptoms including fever, chills, malaise, myalgia, and headache were found to be signifi-
cantly associated with PLA by univariate analysis. Hepatomegaly and right upper quadrant pain were more commonly found in non-PLA patients, but without statistical significance. It might be because more than half of all non-PLA patients had biliary tract infections. Some laboratory parameters, i.e., hemoglobin, total bilirubin, and lactate dehydrogenase, were positively or negatively associated with the presence of PLA by univariate analysis, but only hyperbilirubinemia was negatively associated with PLA by multivariate analysis. Even though such laboratory abnormalities could suggest hepatic abnormalities, these were not uniformly specific for the presence of PLA in patients with community-acquired K. pneumoniae bacteremia. In summary, although several clinical and laboratory findings were significantly associated with PLA in this study, we could not derive consistent clues for the presence of PLA because of their non-specificity. Therefore, in regions where K. pneumoniae is a common cause of bacteremic PLA, abdominal imaging studies should be obtained to rule out PLA in adults without an explainable source of communityacquired K. pneumoniae bacteremia.
Acknowledgements This work was supported by the 2004 Inje University research grant.
371
Letters to the Editor Conflict of interest: No conflict of interest to declare.
9. Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA 1993;270:2957—63.
References 1. Wang JH, Liu YC, Lee SS, Yen MY, Chen YS, Wann SR, et al. Primary liver abscess due to Klebsiella pneumoniae in Taiwan. Clin Infect Dis 1998;26:1434—8. 2. Lee KT, Wong SR, Sheen PC. Pyogenic liver abscess: an audit of 10 years’ experience and analysis of risk factors. Dig Surg 2001;18: 459—65. 3. Cheng HP, Chang FY, Fung CP, Siu LK. Klebsiella pneumoniae liver abscess in Taiwan is not caused by a clonal spread strain. J Microbiol Immunol Infect 2002;35:85—8. 4. Lim SW, Lee EJ, Lee SW, Kim SM, Kim JH, Kim BJ, et al. Clinical significance of Klebsiella pneumoniae in liver abscess. Korean J Gastroenterol 2003;42:226—31. 5. Chang FY, Chou MY. Comparison of pyogenic liver abscesses caused by Klebsiella pneumoniae and non-K. pneumoniae pathogens. J Formos Med Assoc 1995;94:232—7. 6. Tsay RW, Siu LK, Fung CP, Chang FY. Characteristics of bacteremia between community-acquired and nosocomial Klebsiella pneumoniae infection: risk factor for mortality and the impact of capsular serotypes as a herald for community-acquired infection. Arch Intern Med 2002;162: 1021—7. 7. Wang LS, Lee FY, Cheng DL, Liu CY, Hinthorn DR, Jost PM. Klebsiella pneumoniae bacteremia: analysis of 100 episodes. J Formos Med Assoc 1990;89:756—63. 8. Ko WC, Paterson DL, Sagnimeni AJ, Hansen DS, Von Gottberg A, Mohapatra S, et al. Community-acquired Klebsiella pneumoniae bacteremia: global differences in clinical patterns. Emerg Infect Dis 2002;8:160—6.
Recurrent catheter-related bloodstream infections: risk factors and outcome With considerable interest we read the article by Erbay et al., in which the authors examined the risk factors for recurrent catheter-related bloodstream infections (CR-BSI).1 Although Erbay et al. also included general ward patients in their analyses, they found an astonishingly high rate of patients with recurrent CR-BSI.1 In our opinion this enormous incidence, namely one-third of patients, can firstly be explained by their definitions; they define recurrent CR-BSI as positive blood cultures, coupled with positive catheter tip culture ‘or’ no other new source of infection evident. Secondly, it is conceivable that patients admitted to the intensive care unit are more prone to non-infectious signs of systemic inflammation compared to those admitted elsewhere in the hospital. However, we would have liked the authors to detail the indications for which blood sampling for culturing was obtained. CR-BSI should be treated with systemic antimicrobials for 10—14 days.2 In the article by Erbay et al., no data are given concerning the time period in which antimicrobials were administered. Consequently, it may be assumed that some of the CR-BSIs, now categorized as ‘recurrent CR-BSI’, were still related to the first episode. Further, it was mentioned that all patients were given antibiotics; however, no information was provided concerning the time delay to the start of
Sang-Won Park* Department of Internal Medicine, Dankook University College of Medicine, Cheonan, Chungnam 330-715, Korea Baek-Nam Kim Yee-Gyung Kwak Department of Internal Medicine, Inje University College of Medicine, Seoul, Korea Eu-Suk Kim Department of Internal Medicine, Dongguk University College of Medicine, Goyang, Korea *Corresponding author. Tel.: +82 41 550 3918; fax: +82 41 556 3256 Corresponding Editor: Andy I.M. Hoepelman, Utrecht, The Netherlands 26 May 2006 doi:10.1016/j.ijid.2006.09.008
adequate antimicrobials. As appropriateness of empiric antimicrobial therapy within the first hours following onset of infection is of key importance to the improvement of patient outcome, we would be interested in an elaboration of this issue.3—5 CR-BSIs place a significant burden on healthcare resources due to the necessity for additional therapies. The increased morbidity accounts for a significant increase in healthcare expenditure.6 We congratulate Dr Erbay and colleagues for their important work at aiming to identify risk factors related to the development of recurrent CR-BSIs. However, the most important risk factor for CR-BSI is transmission by means of the personnel taking care of the patient. Therefore, in addition to their research, we would like to draw attention to the fact that the cornerstone of CR-BSI prevention is education, with performance feedback and on-the-floor process control following on.7 As CR-BSIs are largely preventable nosocomial infections, an intensified teaching program may heighten awareness among healthcare professionals, so that standards will become routine.8 It has been proven that this is one of the most effective, easiest, and least expensive strategies to reduce CR-BSI rates.9,10 Ultimately, taking full advantage of knowledge of associated risk factors of CR-BSIs, in combination with fitting strategies such as education programs, will lead to reduced antibiotic use and thus alleviation of selection pressure. Conflict of interest: No conflict of interest to declare.
Letters to the Editor nosis) was demonstrated in 63 (38.4%) patients, followed by sialoadenitis in 16 (9.8%) and mucoepidermoid carcinoma in nine (5.5%) patients. MSG tuberculosis was diagnosed in five (3%) patients by either histopathologic study or mycobacterium tuberculosis PCR. Their mean age was 37.4 (range 25—49) years. All patients were female and midwives. The involved MSG was the submandibular gland in four (80%) patients and the parotid gland in only one (20%) patient. None of the patients had a past or family history of TB. Physical examination was unremarkable except for the presence of MSG masses. All patients received standard anti-tuberculous chemotherapy for 8—12 months. Iran is an endemic country for TB, so it is important to have an estimate of TB prevalence in MSG masses. To the best of our knowledge, this is the first formal report in this regard. A recent study in South Korea reported eight cases of MSG tuberculosis among patients admitted at any of the three tertiary referral centers in a period of 10 years.11 The parotid and submandibular glands were involved in five (62.5%) and three cases (37.5%), respectively. Another study found only one case of salivary gland TB among 150 patients with salivary gland disease in a 14-year experience in India.12 TB involvement was located in a minor salivary gland in that study. In another study of 155 salivary gland masses, 10 (6.4%) cases of TB infection were found.13 A study in Turkey reported parotid involvement of TB in six cases among 216 patients with a parotid mass.14 Finally, in a 12-year study in China, nine patients were reported to have suffered from tuberculosis of the parotid glands.15 The prevalence of MSG tuberculosis in our study is relatively similar to that in other reports from Asia. Considering the extremely rare event of tuberculous involvement of the submandibular gland,3,4 our experience is unusual in that 80% of our patients had their submandibular glands involved. We doubt that this unusually high rate of submandibular gland tuberculosis could be regarded as incidental. However, we are at present unclear as to the mechanism of TB’s strong tendency to involve the submandibular glands in our patients. Studies with larger sample sizes are needed to clarify the significance of this finding. Conflict of interest: No conflict of interest to declare.
References 1. Choudhury N, Bruch G, Kothari P, Rao G, Simo R. 4 years’ experience of head and neck tuberculosis in a south London hospital. J R Soc Med 2005;98:267—9. 2. Kundu S, Das S, Dey A, Sengupta A. Tuberculosis of parotid gland–— a rare clinical entity. Indian J Otolaryngol Head Neck Surg 2004;56:57—8.
Clinical factors indicating the presence of pyogenic liver abscess in community-acquired Klebsiella pneumoniae bacteremia Klebsiella pneumoniae is a leading cause of pyogenic liver abscess (PLA) in some Asian countries,1—5 where one of the most common sources of K. pneumoniae bacteremia is PLA.6—8
3. Manipoud P, Rerolle S, Kemeny JL, Lafaye M. Tuberculosis of the submandibular gland. Apropos of a case and review of the literature. Ann Otolaryngol Chir Cervicofac 1993;110:482—4. 4. Bodner L, Lewin-Epstein J, Shteyer A. Submandibular tuberculous lymphadenitis (scrofula): report of two cases. J Oral Maxillofac Surg 1990;48:192—6. 5. Mignogna MD, Muzio LL, Favia G, Ruoppo E, Sammartino G, Zarrelli C, et al. Oral tuberculosis: a clinical evaluation of 42 cases. Oral Dis 2000;6:25—30. 6. Rowe-Jones JM, Vowles R, Lighton SE, Freedman AR. Diffuse tuberculous parotitis. J Laryngol Otol 1992;106:1094—5. 7. Tuli AS, Gupta V, Singh H, Chary G, Chand AK. Primary tuberculosis of parotid gland. Indian J Otolaryngol Head Neck Surg 2005;57:82—3. 8. Suleiman AM. Tuberculous parotitis: report of 3 cases. Br J Oral Maxillofac Surg 2001;39:320—3. 9. Lee IK, Liu JW. Tuberculous parotitis: case report and literature review. Ann Otol Rhinol Laryngol 2005;114:547—51. 10. Hamdan AL, Hadi U, Shabb N. Tuberculous parotitis: a forgotten entity. Otolaryngol Head Neck Surg 2002;126:581—2. 11. Kim YH, Jeong WJ, Jung KY, Sung MW, Kim KH, Kim CS. Diagnosis of major salivary gland tuberculosis: experience of eight cases and review of the literature. Acta Otolaryngol 2005;125:1318—22. 12. Tewari M, Shukla HS, Kumar M, Sharma OP. Non-neoplastic salivary gland disease with reference to minor salivary gland tuberculosis. Indian J Surg 2003;65:168—71. 13. Shaha AR, Webber C, DiMaio T, Jaffe BM. Needle aspiration biopsy in salivary gland lesions. Am J Surg 1990;160:373—6. 14. Suoglu Y, Erdamar B, Colhan I, Katircioglu OS, Cevikbas U. Tuberculosis of the parotid gland. J Laryngol Otol 1998;112: 588—91. 15. Chou YH, Tiu CM, Liu CY, Hong TM, Lin CZ, Chiou HJ, et al. Tuberculosis of the parotid gland. J Ultrasound Med 2004;23: 1275—81.
Pantea Hashemi Armin Rashidi* Ebrahim Razmpa Department of Otolaryngology, Imam Khomeini Hospital, Tehran University of Medical Sciences, Tehran, Iran *Corresponding author. Address: No 12, Aramesh Alley, Mellat Park, Tehran, Iran. Tel.: +98 21 22042595 E-mail address: [email protected] (A. Rashidi) Corresponding Editor: Michael Whitby, Brisbane, Australia 17 July 2006 doi:10.1016/j.ijid.2006.09.006
PLA presents a diagnostic challenge, because half of all patients show no specific symptoms or signs that direct attention to the liver. This study was conducted to identify clinical and laboratory factors indicating the early diagnosis of PLA in community-acquired K. pneumoniae bacteremia. Patients at three hospitals in Korea, diagnosed with K. pneumoniae bacteremia between January 2001 and December 2004, were retrospectively reviewed with regard to
370 Table 1
Letters to the Editor Clinical characteristics of patients with community-acquired Klebsiella pneumoniae bacteremiaa
Variables
PLA group (N = 38)
Non-PLA group (N = 93)
p Value
Age (mean standard deviation) years Male:female
62.1 12.9 1.2
62.2 13.8 1.3
1.0 0.9
Underlying conditions Pancreaticobiliary disease including tumor Liver cirrhosis with/without hepatoma Diabetes mellitus None
7 (18.4%) 0 9 (23.7%) 25 (65.8%)
27 12 25 41
0.2 0.02 0.7 0.04
Symptoms Right upper quadrant pain
9 (23.7%)
28 (30.1%)
0.5
Physical findings Body temperature >38 or <36 8C Hepatomegaly
24 (63.2%) 1 (2.6%)
54 (58.1%) 4 (4.3%)
0.6 1.0
Laboratory findings White cell count >12 or <4 109/l Platelet <100 109/l C-reactive protein >10 mg/dl Alanine transaminase >100 U/L Albumin <3 g/dl Total bilirubin >1.5 mg/dl Alkaline phosphatase >300 U/L Lactate dehydrogenase >400 U/L Chest X-ray abnormalities b
19 (50.0%) 16 (42.1%) 28/35 (80.0%) 17 (44.7%) 15/37 (40.5%) 14 (36.8%) 3/37 (8.1%) 22/27 (81.5%) 6 (15.8%)
45 (48.4%) 33 (35.5%) 49/74 (66.2%) 34 (36.6%) 39/90 (43.3%) 58 (62.4%) 8/89 (9.0%) 32/61 (52.5%) 13 (14.0%)
0.9 0.5 0.1 0.4 0.8 0.008 1.0 0.01 0.8
Severity of illness SAPS II score of 30 Septic shock
8 (21.1%) 5 (13.2%)
24 (25.8%) 12 (12.9%)
0.57 1.0
(29.0%) (12.9%) (26.9%) (44.1%)
SAPS, simplified acute physiology score. a Not all parameters are presented. b Chest X-ray abnormalities in the right lower lung field included pleural effusion, atelectasis, or elevated right hemidiaphragm.
demographic, clinical, and laboratory data. Healthcare-associated and nosocomial cases were excluded because K. pneumoniae bacteremia associated PLA is usually community-acquired.1 Only community-acquired cases in adults who had undergone abdominal ultrasonography (US) or computed tomography (CT) were included. The severity of acute illness was assessed at the time of initial bacteremia by using the simplified acute physiology score (SAPS) II.9 Among a total of 311 adult patients with K. pneumoniae bacteremia, 131 met the inclusion criteria and were analyzed. In 38 of the 131 patients, K. pneumoniae bacteremia was associated with PLA (PLA group). PLA was cryptogenic in 31 patients and of biliary origin in seven. In 93 patients (non-PLA group), the sources of bacteremia were biliary tract (49), unknown source (17), urinary tract (16), lung (5), peritoneum (3), bone and joint (2), and skin and soft tissue (1). Bacteremia was polymicrobial in five patients only, in the non-PLA group ( p = 0.32). Clinical characteristics are compared in Table 1. Multiple logistic regression revealed that only hyperbilirubinemia (total bilirubin >1.5 mg/dl) was negatively associated with PLA (odds ratio 0.16; 95% confidence interval 0.04—0.73; p = 0.02). Diabetes mellitus and pancreaticobiliary disease were the most common underlying conditions in our patients, but no significant difference was found between the PLA group and nonPLA group. Several symptoms including fever, chills, malaise, myalgia, and headache were found to be signifi-
cantly associated with PLA by univariate analysis. Hepatomegaly and right upper quadrant pain were more commonly found in non-PLA patients, but without statistical significance. It might be because more than half of all non-PLA patients had biliary tract infections. Some laboratory parameters, i.e., hemoglobin, total bilirubin, and lactate dehydrogenase, were positively or negatively associated with the presence of PLA by univariate analysis, but only hyperbilirubinemia was negatively associated with PLA by multivariate analysis. Even though such laboratory abnormalities could suggest hepatic abnormalities, these were not uniformly specific for the presence of PLA in patients with community-acquired K. pneumoniae bacteremia. In summary, although several clinical and laboratory findings were significantly associated with PLA in this study, we could not derive consistent clues for the presence of PLA because of their non-specificity. Therefore, in regions where K. pneumoniae is a common cause of bacteremic PLA, abdominal imaging studies should be obtained to rule out PLA in adults without an explainable source of communityacquired K. pneumoniae bacteremia.
Acknowledgements This work was supported by the 2004 Inje University research grant.
371
Letters to the Editor Conflict of interest: No conflict of interest to declare.
9. Le Gall JR, Lemeshow S, Saulnier F. A new simplified acute physiology score (SAPS II) based on a European/North American multicenter study. JAMA 1993;270:2957—63.
References 1. Wang JH, Liu YC, Lee SS, Yen MY, Chen YS, Wann SR, et al. Primary liver abscess due to Klebsiella pneumoniae in Taiwan. Clin Infect Dis 1998;26:1434—8. 2. Lee KT, Wong SR, Sheen PC. Pyogenic liver abscess: an audit of 10 years’ experience and analysis of risk factors. Dig Surg 2001;18: 459—65. 3. Cheng HP, Chang FY, Fung CP, Siu LK. Klebsiella pneumoniae liver abscess in Taiwan is not caused by a clonal spread strain. J Microbiol Immunol Infect 2002;35:85—8. 4. Lim SW, Lee EJ, Lee SW, Kim SM, Kim JH, Kim BJ, et al. Clinical significance of Klebsiella pneumoniae in liver abscess. Korean J Gastroenterol 2003;42:226—31. 5. Chang FY, Chou MY. Comparison of pyogenic liver abscesses caused by Klebsiella pneumoniae and non-K. pneumoniae pathogens. J Formos Med Assoc 1995;94:232—7. 6. Tsay RW, Siu LK, Fung CP, Chang FY. Characteristics of bacteremia between community-acquired and nosocomial Klebsiella pneumoniae infection: risk factor for mortality and the impact of capsular serotypes as a herald for community-acquired infection. Arch Intern Med 2002;162: 1021—7. 7. Wang LS, Lee FY, Cheng DL, Liu CY, Hinthorn DR, Jost PM. Klebsiella pneumoniae bacteremia: analysis of 100 episodes. J Formos Med Assoc 1990;89:756—63. 8. Ko WC, Paterson DL, Sagnimeni AJ, Hansen DS, Von Gottberg A, Mohapatra S, et al. Community-acquired Klebsiella pneumoniae bacteremia: global differences in clinical patterns. Emerg Infect Dis 2002;8:160—6.
Recurrent catheter-related bloodstream infections: risk factors and outcome With considerable interest we read the article by Erbay et al., in which the authors examined the risk factors for recurrent catheter-related bloodstream infections (CR-BSI).1 Although Erbay et al. also included general ward patients in their analyses, they found an astonishingly high rate of patients with recurrent CR-BSI.1 In our opinion this enormous incidence, namely one-third of patients, can firstly be explained by their definitions; they define recurrent CR-BSI as positive blood cultures, coupled with positive catheter tip culture ‘or’ no other new source of infection evident. Secondly, it is conceivable that patients admitted to the intensive care unit are more prone to non-infectious signs of systemic inflammation compared to those admitted elsewhere in the hospital. However, we would have liked the authors to detail the indications for which blood sampling for culturing was obtained. CR-BSI should be treated with systemic antimicrobials for 10—14 days.2 In the article by Erbay et al., no data are given concerning the time period in which antimicrobials were administered. Consequently, it may be assumed that some of the CR-BSIs, now categorized as ‘recurrent CR-BSI’, were still related to the first episode. Further, it was mentioned that all patients were given antibiotics; however, no information was provided concerning the time delay to the start of
Sang-Won Park* Department of Internal Medicine, Dankook University College of Medicine, Cheonan, Chungnam 330-715, Korea Baek-Nam Kim Yee-Gyung Kwak Department of Internal Medicine, Inje University College of Medicine, Seoul, Korea Eu-Suk Kim Department of Internal Medicine, Dongguk University College of Medicine, Goyang, Korea *Corresponding author. Tel.: +82 41 550 3918; fax: +82 41 556 3256 Corresponding Editor: Andy I.M. Hoepelman, Utrecht, The Netherlands 26 May 2006 doi:10.1016/j.ijid.2006.09.008
adequate antimicrobials. As appropriateness of empiric antimicrobial therapy within the first hours following onset of infection is of key importance to the improvement of patient outcome, we would be interested in an elaboration of this issue.3—5 CR-BSIs place a significant burden on healthcare resources due to the necessity for additional therapies. The increased morbidity accounts for a significant increase in healthcare expenditure.6 We congratulate Dr Erbay and colleagues for their important work at aiming to identify risk factors related to the development of recurrent CR-BSIs. However, the most important risk factor for CR-BSI is transmission by means of the personnel taking care of the patient. Therefore, in addition to their research, we would like to draw attention to the fact that the cornerstone of CR-BSI prevention is education, with performance feedback and on-the-floor process control following on.7 As CR-BSIs are largely preventable nosocomial infections, an intensified teaching program may heighten awareness among healthcare professionals, so that standards will become routine.8 It has been proven that this is one of the most effective, easiest, and least expensive strategies to reduce CR-BSI rates.9,10 Ultimately, taking full advantage of knowledge of associated risk factors of CR-BSIs, in combination with fitting strategies such as education programs, will lead to reduced antibiotic use and thus alleviation of selection pressure. Conflict of interest: No conflict of interest to declare.