Int J Clin Exp Med 2015;8(10):19775-19780 www.ijcem.com /ISSN:1940-5901/IJCEM0013695
Original Article Prevention options for ventriculoperitoneal shunt infections: a retrospective analysis during a five-year period Xing Wu1, Qin Liu2, Xiaofei Jiang2, Tao Zhang2 Department of Neurosurgery, Huashan Hospital, Fudan University, Shanghai, China; 2Department of Laboratory Medicine, Huashan Hospital, Fudan University, 12 Central Urumqi Road, Shanghai 200040, China 1
Received July 30, 2015; Accepted September 28, 2015; Epub October 15, 2015; Published October 30, 2015 Abstract: Shunt infection is the most common and dreaded complication in patients with hydrocephalus. For the purpose of reducing the high morbidity and mortality, how to prevent is a vital step. A retrospective analysis of 384 CSF cerebrospinal fluid (CSF) shunt procedures was undertaken from 2006 to 2010 in our neurosurgery department. Infection diagnosis was established by subjecting the CSF to biochemical and microbiological parameters. The patients’ demographic and clinical characteristics, various treatment procedures and outcome were evaluated. The infection rate of ventriculoperitoneal (VP) shunt was 12.5% in 2006, which dropped to 2% and stabilized at lower level from 2008. The most common causes of hydrocephalus were traumatic injury and brain tumor. Fever and consciousness disturbance were the major clinical symptoms. Gram-negative rods episodes was the most frequently isolated microorganisms accounting for 58%, followed by S. aureus , S. epidermidis and Staphylococcus haemolyticus. With the removal of shunt and intravenous antibiotics therapy, 82% of the patients survived. Majority of the isolates were sensitive to the carbopenem antibiotics and vancomycin. The mean length of hospital stay was 47 days. Prompt shunt removal and perioperative antibiotic prophylaxis seems to be essential for the survival of patients with VP shunt infection. Keywords: VP shunt infection, CSF, perioperative antibiotic prophylaxis, vancomycin
Introduction Any variation in the production or sorption of CSF leads to hydrocephalus, such as aqueductal stenosis, spina bifida, intraventricular hemorrhage, meningitis, tumors, traumatic head injury, or subarachnoid hemorrhage. Surgical diversion of hydrocephalus via a ventriculoperitoneal (VP) shunt is the preferred method of treatment in most patients. Infection was the most common shunt complication in patients with hydrocephalus, which occurs with a variable rate of 2.2-39% [1, 2]. Several risk factors have been recognized, including the age of patient, the type of shunt implanted, duration of the shunt placement operation, the experience of the neurosurgeon, and the cause of hydrocephalus [3-5]. Shunt associated infections are most frequently caused by skin bacteria, such as CoNS and Staphylococcus aureus.
Gram negative bacteria are the next most frequent pathogens, responsible for 19% to 22% of cases [6, 7]. Despite the emergence of modern neurosurgical technology and new antibodies, infection after VP shunt remains enormously higher than that in other neurosurgical operations that involve implanted device. Even when patients obtain appropriate management strategies, VP infections are related to a vital risk of morbidity, including haryencephalia, seizure, and neurological deficit [8]. We sought to retrospect and analyze systematically the epidemiological, clinical, laboratory, and microbiological characteristics of VP shunt associated infections and treatment outcome collected from 2006 to 2010 shunt infection in our neurosurgery department. The objectives of this study were to evaluate the infection rate and mortality associated with VP shunt inser-
Prevention options for VP shunt infections organisms seen on CSF Gram stain, or organisms found in blood culture. Excluded from the study were patients with ventriculoatrial shunt, lumboperitoneal shunt, pre-existent CSF infection, incomplete infection and microbiological data, and patients whose shunt was removed before a diagnosis of infection was determined. CSF was obtained through needle aspiration of the shunt reservoir.
Table 1. Demographic and epidemiological characteristics of patients with VP shunt insertion
Age (median and range) Gender Male Female Reason for shunt insertion Head trauma Brain tumor Intracranial hemorrahage Other Length of hospital stay *
VP infection (%) Non-infection (%) (n = 28) (n = 356) 38 (6-76) 43 (5-82) 25 (89)* 3 (11)
220 (62) 136 (38)
11 (39) 9 (32) 5 (18) 3 (11) 47 (6-148)*
51 (14) 165 (46) 38 (11) 102 (29) 22 (5-271)
P < 0.05.
tion and to assess what factors, if any, were the essential steps in the prevention of VP shunt infection. Patient and method A retrospective chart review was performed of all patients aged 2 years or older submitted to VP shunt surgery between January 2006 and July 2010 at the Neurosurgery Department of Huashan Hospital, Fudan University, which is the largest neurosurgical medical center in Shanghai, China, where, collectively, approximately 8,000 neurosurgical procedures are performed annually. Clinical and laboratory data, the clinical condition requiring shunt procedure, as well as information on response to treatment and outcome were obtained from each patient’s medical records. As preoperative antibiotic prophylaxis, a single dose of vancomycin was routinely used before 2007. It generally consisted of the third-generation cephalosporin and vancomycin administered prior to skin incision from 2008. To define the shunt infected group, the criteria used in this study were established according to previous studies [9, 10]. VP shunt infection was defined as at least one of the following criteria was present: (1) isolation of the organism from 1 or more cerebrospinal fluid (CSF) cultures or (2) fever (temperature > 38°C), headache, neck stiffness, cranial nerve signs, or irritability without another recognized cause; a typical CSF findings such as: CSF leukocyte count > 250×106 cells/L, CSF total protein > 450 mg/L, CSF glucose level < 2.5 mmol/L,
19776
Statistical analysis Statistical analysis was performed using the SPSS 10.0 software package (Version 11.5; SPSS, Chicago, IL, USA). The variables are reported as absolute and relative frequencies and as the mean and standard deviation. Quantitative variables were analyzed with the Student t test or the Mann-Whitney test when appropriate. Qualitative variables were analyzed with the chi-square test with the Yates correction or the Fisher exact test (2-tailed) when necessary. A P value less than .05 was considered statistically significant. Results During the study period, 28 VP shunt infections resulted from 384 shunt procedures, giving an overall infection rate of 7.3%. The infection rate of VP shunt was 12.5% (12/96) and 10% (12/97) in 2006 and 2007, which significantly dropped to 2% and stabilized at lower level from 2008. The 28 patients with VP shunt infections including 3 pediatric patients and 25 adult patients (mean age: 38 years old) (Table 1). 89% of infections (25 episodes) involved male patients. The duration of symptoms before diagnosis occurred within 6 months of the procedure, ranging from 1 day to 158 days (mean, 18.8 days). Of the 28 infections, up to two-thirds (85%), the shunt infection occurred within 30 days post-operation, which was either the primary shunt insertions or shunt revisions. Late infections occurred in 4 cases, about two months, three months and five months after shunt revision, respectively. The clinical features of shunt infection were similar in children and adult (Table 2). Nearly
Int J Clin Exp Med 2015;8(10):19775-19780
Prevention options for VP shunt infections Table 2. Clinical features of VP shunt infection of 28 patients Parameters Clinical symptom Temperature > 38°C Consciousness disturbances Neck stiffness Headache Infection signal by CT scan Associated infection Pneumonia Intra-abdominal infection Urinary tract infection Underlying diseases Diabetes Cardiovascular disease Tuberculosis Hypokalemia CSF Leucocytes > 250×106 Glucose level were < 2.5 mmol/L Protein > 450 mg/L Time between shunt insertion and diagnosis of infection < 30 days > 30 days
all patients (96%) presented with fever > 38.0°C. Although consciousness disturbances was noted in 71% of cases, only 9 (32%) of 28 cases was found neck stiffness, and 21% of cases complained of headache. The cranial CT scan showed signs of shunt infection in 3 (11%) of 28 patients. Pneumonia was the most frequent associated infection (39%), followed for intra-abdominal infection (18%), and urinary tract infection (14%). In addition, 9 (32%) patients had other severe underlying diseases such as diabetes, cardiovascular disease, tuberculosis or hypokalemia. CSF studies revealed leukocytosis, elevated total protein and decreased glucose levels. CSF leucocytes were more than 250×106 in 75% of shunt infections, and 21 of 28 (75%) CSF samples displayed neutrophil predominance. CSF total protein levels were abnormal in 57% (16 of 28), with a median of 982.48±743.91 mg/L (range, 106±3530 mg/d). CSF glucose level were < 2.5 mmol/L in 23 patients (82%), with a median of 1.98±0.86 mmol/L (range, 0.56-3.6 mmol/L). The pathogens causing shunt infection were determined in 26 episodes (93%, Table 3). In 19777
No. of patients (%) 27 (96) 20 (71) 9 (32) 6 (21) 3 (11) 11 (39) 5 (18) 4 (14) 4 (14) 3 (11) 1 (4) 1 (4)
two patients, no causative pathogen was identified. The most prevalent organisms were gram-negative rods episodes (15), S. aureus (4), S. epidermidis (3), Staphylococcus haemolyticus (2), coagulase-negative staphylococci (1), enterococcus faecium (1). Of the 15 gramnegative rods episodes, 5 (33%) were due to acinetobacter baumannii, 4 (27%) were due to klebsiella pneumoniae, 2 (13%) were due to Pseudomonas aeruginosa and Pseudomonas maltophilia, respectively, and one each by enterobacter cloacae and ent. agglomerans.
All the patients received appropriate antibiotic treatments. VP 21 (75) devices were promptly removed from the 15 patients. In 13 epi23 (82) sodes (46%), neurosurgical de16 (57) vices was not removed because of good response to antibiotic 24 (86) treatment alone (10 episodes), 4 (14) or the deterioration of the patient’s complications (3 episodes). When the diagnosis of gram-negative rods meningitis was established, carbopenem antibiotics (meropenem or imipenem) was used in 9 (64%) of 14 cases, with or without amikacin; the rest of cases were treated with a thirdgeneration cephalosporin and rifampin. Five patients with Staphylococcus aureus and coagulase-negative staphylococci infection were treated with vancomycin. In addition, 3 patients received intrathecal therapy with gentamicin and vancomycin. The response was successful in 23 cases. Overall mortality was 18% (5/28). Three of the five expired patients died in septic shock, whereas the remaining patient died from disseminated intravascular coagulation and multiple organ dysfunction syndromes. Discussion Shunt-associated infection is a severe complication of VP shunt therapy for hydrocephalus. Despite the appropriate care received by patients, VP shunt infection remains a major cause of morbidity and even mortality [11, 12]. An infection rate of 6.98% was observed in the present study. Multiple factors have been Int J Clin Exp Med 2015;8(10):19775-19780
Prevention options for VP shunt infections Table 3. Summary of clinical data in 28 patients with VP shunt infection Case 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Shunt removal No Yes Yes Yes No Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes No Yes No Yes Yes Yes Yes Yes Yes No Yes
Causative pathogen klebsiella pneumoniae klebsiella pneumoniae klebsiella pneumoniae S. epidermidis S. aureus Pseudomonas aeruginosa Pseudomonas maltophilia acinetobacter baumannii enterobacter cloacae S. aureus acinetobacter baumannii acinetobacter baumannii acinetobacter baumannii coagulase-negative staphylococci S. epidermidis staphylococcus haemolyticus S. aureus Pseudomonas maltophilia Pseudomonas aeruginosa negtive enterococcus faecium staphylococcus haemolyticus S. epidermidis ent.agglomerans klebsiella pneumoniae S. aureus negtive acinetobacter baumannii
brought up as predisposing to infection after VP shunt, in particular the age at shunt insertion, cause of hydrocephalus, the experience of the neurosurgeon, the cause of hydrocephalus, operative room traffic and perioperative antibiotics [4-6]. In accordance with other studies, the majority of shunt-associated infections in the present study occurred within 30 days after shunt insertions. Fever and consciousness disturbances were the primary clinical symptoms among the patients, whereas typical neurological signs-such as neck stiffness or headache-were present in less than one-half of the episodes. This phenomenon reveals the characteristics of implant-associated infections are typically caused by low-virulence and slowly replicating organisms growing on the shunt surface in biofilms [13]. Early detection of shunt infection is usually difficult because of its nonspecific clinical signs, once patients 19778
Antibody treatment (mg/kg/day) Imipenem Amikacin Meropenem Meropenem Vancomycin Teicoplanin Vancomycin Teicoplanin Rifampin Ceftazidime Imipenem Amikacin Meropenem Vancomycin Amikacin Fosfomycin Meropenem Ceftazidime Vancomycin Gentamicin Vancomycin Vancomycin Fosfomycin Vancomycin Teicoplanin Rifampin Imipenem Ceftazidime Gentamicin Vancomycin Amikacin Teicoplanin Imipenem Amikacin Meropenem Vancomycin Vancomycin Imipenem Amikacin
Intrathecal therapy No Yes No Yes No No No No No No No Yes No Yes No No No No No No Yes No No No No Yes No No
Outcomes Died Survival Survival Survival Died Survival Survival Survival Survival Survival Died Survival Survival Survival Survival Survival Survival Died Survival Survival Survival Died Survival Survival Survival Survival Survival Survival
with fever and delayed recovery of consciousness after a VP procedure, the occurrence of central nervous system infection should be suspected. The main causative agents of shunt infections are gram-negative rods episodes in our study. This is contrary to other studies, which have shown the most common episodes to be grampositive bacterium [4, 6, 7]. The infection rate of gram-negative bacterium was higher relatively in our study, it can be speculated that the routine use of antacids leads to bacterial colonization of the stomach with Gram-negative strains may figure out why a high percentage of the infection are caused by gram-negative rods in our study [14, 15]. In a recent study reported in our hospital the results are similar to ours that the most freInt J Clin Exp Med 2015;8(10):19775-19780
Prevention options for VP shunt infections quent isolated organism of nosocomial infections is Gram-negative microorganisms. This finding supports the contention that simultaneous infections in other parts of the body which were caused by the same gram negative microorganisms may be relevant to that higher incidence. S. epidermidis, S. aureus, and Staphylococcus haemolyticus were the most frequent isolate encountered in our gram-positive series. Some patients have received intrathecal therapy with amikacin and vancomycin. A successful outcome was observed in 3 cases receiving intrathecal therapy, but there were not enough cases to draw reliable conclusions. The goal of antibiotic prophylaxis in shunt surgery is to prevent the shunt hardware with organisms, with particular lay stress on Grampositive bacteria. Historically, the organisms most frequently causing infections of VP shunt are the Staphylococcus species. The causative episodes are coagulase-negative staphylococci in 80% of cases, and most of the organisms isolated are methicillin resistant Staphylococcus epidermidis (MRSE) [16-19]. Antibiotics appropriate for prophylaxis is based on the above-mentioned literature, vancomycin was as used routinely antibody prophylaxis theory. However, the infection rate of VP shunt was higher (12.4% and 10%) in 2006 and 2007. When administrated alone, intra-ventricular vancomycin was associated with the increasing Gram-positive bacteria shunt infection. Thus, clinicians consider the initial antibiotic treatment with both Norvancomycin and Ceftriaxone may provide the satisfactory prophylactic effect. After this change in protocol, only 7 infections occurred in the next year and 2 cases found after that. VP shunt infections was broken down into 2% and stabilized at lower level from 2008. Of note, short-term perioperative antibiotic prophylaxis may be of benefit in preventing shunt infections. Despite satisfactory sensitivity of antimicrobial therapy, an overall mortality of 18% was observed. In postoperative cases, the removal of the infected CSF shunt is considered one of the most important parts of therapy [4, 6, 8]. We found that CSF shunt removal was associated with a lower mortality rate in patients with VP shunt-related infections, and, therefore, the diagnosis of bacterial meningitis in patients with CSF shunts should remind the neurosurgeon to remove foreign body devices. 19779
Conclusion In conclusion, although gram positive organisms were the most frequently seen nosocomial VP bacterial infection in neurosurgical patients, our data show that infections with gram-negative bacteria form a significant episodes percentage in the shunt infection in our center. In the absence of controlled trials, the result of this study may suggest that management option of infected shunts, which consists of perioperative antibiotic prophylaxis (in our survey vancomycin and a third generation cephalosporin) plus removal of the infected shunt, seems to be essential for the survival of patients with VP shunt infection. Acknowledgements This work was supported by the National Clinical Key Subject, the National Natural Science Foundation of China (grant number 81072070, 81471241, 81171133, and 81271375) and the Natural Science Foundation of Shanghai (grant number 15ZR1405600). Disclosure of conflict of interest None. Address correspondence to: Dr. Tao Zhang, Department of Laboratory Medicine, Huashan Hospital, Fudan University, 12 Central Urumqi Road, Shanghai 200040, China. E-mail: [email protected]
References [1]
[2]
[3]
[4]
[5]
Pople IK, Bayston R, and Hayward RD. Infection of cerebrospinal fluid shunts in infants: a study of etiological factors. J Neurosurg 1992; 77: 29-36. Drake JM, Kestle JR, Milner R, Cinalli G, Boop F, Piatt J Jr, Haines S, Schiff SJ, Cochrane DD, Steinbok P, MacNeil N. Randomized trial of cerebrospinal fluid shunt valve design in pediatric hydrocephalus. Neurosurg 1998; 43: 294-303 Shunts. In Infections Associated with Indwelling Medical Devices. In: Bisno AL, Waldvogel FA, editors. Washington: American Society for Microbiology; 1994. pp. 91-109. Piedra MP, Ragel BT, Dogan A, Coppa ND and Delashaw JB. Timing of cranioplasty after decompressive craniectomy for ischemic or hemorrhagic stroke. J Neurosurg 2013; 118: 109114 Reddy GK, Bollam P, and Caldito G. Ventriculoperitoneal shunt surgery and the risk of
Int J Clin Exp Med 2015;8(10):19775-19780
Prevention options for VP shunt infections shunt infection in patients with hydrocephalus: long-term single institution experience. World Neurosurg 2012; 8: 155-163. [6] Dallacasa P, Dappozzo A, Galassi E, Sandri F, Cocchi G and Masi M. Cerebrospinal fluid shunt infections in infants. Childs Nerv Syst 1995; 11: 643-649. [7] Turgut M, Alabaz D, Erbey F, Kocabas E, Erman T, Alhan E and Aksaray N. Cerebrospinal fluid shunt infections in children. Pediatr Neurosurg 2005; 41: 131-136. [8] Fan-Havard P, and Nahata MC. Treatment and prevention of infections of cerebrospinal fluid shunts. Clin Pharm 1987; 6: 866-880. [9] Wang KW, Chang WN, Shih TY, Huang CR, Tsai NW, Chang CS, Chuang YC, Liliang PC, Su TM, Rau CS, Tsai YD, Cheng BC, Hung PL, Chang CJ and Lu CH. Infection of cerebrospinal fluid shunts: causative pathogens, clinical features, and outcomes. Jpn JInfect Dis 2004; 57: 4448. [10] Garner JS, Jarvis WR, Emori TG, Horan TC and Hughes JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control 1988; 116: 128-140. [11] Tuan TJ, Thorell EA, Hamblett NM, Kestle JR, Rosenfeld M, and Simon TD. Treatment and microbiology of repeated cerebrospinal fluid shunt infections in children. Pediatr Infect Dis 2011; 30: 731-735. [12] Warf BC, Bhai S, Kulkarni AV and Mugamba J. Shunt survival after failed endoscopic treatment of hydrocephalus. J Neurosurg Pediatr 2012; 10: 463-470.
19780
[13] Conen A, Walti LN, Merlo A, Fluckiger U, Battegay M and Trampuz A. Characteristics and treatment outcome of cerebrospinal fluid shunt-associated infections in adults: a retrospective analysis over an 11-year period. Clin Infect Dis 2008; 47: 73-82. [14] Aubert G, Jacquemond G, Pozzetto B, Duthel R, Boylot D, Brunon J and Dorche G. Pharmacokinetic evidence of imipenem efficacy in the treatment of Klebsiella pneumoniae nosocomial meningitis. J Antimicrob Chemother 1991; 28: 316-317. [15] Vallée E, Joly-Guillou ML, and BergogneBerezin E. Comparative activity of imipenem, ceftazidime and cefotaxime against Acinetobacter calcoaceticus. Presse Med 1990; 19: 588-591. [16] Mullan E, Lucas C, Mackie S and Carachi R. Audit of ventriculoperitoneal shunt infections in paediatric patients, 2006-2013. Scott Med J 2014; 59: 198-203. [17] Demetriades AK, and Bassi S. Antibiotic resistant infections with antibiotic-impregnated Bactiseal catheters for ventriculoperitoneal shunts. Br J Neurosurg 2011; 25: 671-673. [18] Bayston R, Ullas G, and Ashraf W. Action of linezolid or vancomycin on biofilms in ventriculoperitoneal shunts in vitro. Antimicrob Agents Chemother 2012; 6: 2842-2845. [19] Tomio R, Akiyama T, Shibao S and Yoshida K. Procalcitonin as an early diagnostic marker for ventriculoperitoneal shunt infections. Surg Infect (Larchmt) 2013; 14: 433-436.
Int J Clin Exp Med 2015;8(10):19775-19780
Original Article Prevention options for ventriculoperitoneal shunt infections: a retrospective analysis during a five-year period Xing Wu1, Qin Liu2, Xiaofei Jiang2, Tao Zhang2 Department of Neurosurgery, Huashan Hospital, Fudan University, Shanghai, China; 2Department of Laboratory Medicine, Huashan Hospital, Fudan University, 12 Central Urumqi Road, Shanghai 200040, China 1
Received July 30, 2015; Accepted September 28, 2015; Epub October 15, 2015; Published October 30, 2015 Abstract: Shunt infection is the most common and dreaded complication in patients with hydrocephalus. For the purpose of reducing the high morbidity and mortality, how to prevent is a vital step. A retrospective analysis of 384 CSF cerebrospinal fluid (CSF) shunt procedures was undertaken from 2006 to 2010 in our neurosurgery department. Infection diagnosis was established by subjecting the CSF to biochemical and microbiological parameters. The patients’ demographic and clinical characteristics, various treatment procedures and outcome were evaluated. The infection rate of ventriculoperitoneal (VP) shunt was 12.5% in 2006, which dropped to 2% and stabilized at lower level from 2008. The most common causes of hydrocephalus were traumatic injury and brain tumor. Fever and consciousness disturbance were the major clinical symptoms. Gram-negative rods episodes was the most frequently isolated microorganisms accounting for 58%, followed by S. aureus , S. epidermidis and Staphylococcus haemolyticus. With the removal of shunt and intravenous antibiotics therapy, 82% of the patients survived. Majority of the isolates were sensitive to the carbopenem antibiotics and vancomycin. The mean length of hospital stay was 47 days. Prompt shunt removal and perioperative antibiotic prophylaxis seems to be essential for the survival of patients with VP shunt infection. Keywords: VP shunt infection, CSF, perioperative antibiotic prophylaxis, vancomycin
Introduction Any variation in the production or sorption of CSF leads to hydrocephalus, such as aqueductal stenosis, spina bifida, intraventricular hemorrhage, meningitis, tumors, traumatic head injury, or subarachnoid hemorrhage. Surgical diversion of hydrocephalus via a ventriculoperitoneal (VP) shunt is the preferred method of treatment in most patients. Infection was the most common shunt complication in patients with hydrocephalus, which occurs with a variable rate of 2.2-39% [1, 2]. Several risk factors have been recognized, including the age of patient, the type of shunt implanted, duration of the shunt placement operation, the experience of the neurosurgeon, and the cause of hydrocephalus [3-5]. Shunt associated infections are most frequently caused by skin bacteria, such as CoNS and Staphylococcus aureus.
Gram negative bacteria are the next most frequent pathogens, responsible for 19% to 22% of cases [6, 7]. Despite the emergence of modern neurosurgical technology and new antibodies, infection after VP shunt remains enormously higher than that in other neurosurgical operations that involve implanted device. Even when patients obtain appropriate management strategies, VP infections are related to a vital risk of morbidity, including haryencephalia, seizure, and neurological deficit [8]. We sought to retrospect and analyze systematically the epidemiological, clinical, laboratory, and microbiological characteristics of VP shunt associated infections and treatment outcome collected from 2006 to 2010 shunt infection in our neurosurgery department. The objectives of this study were to evaluate the infection rate and mortality associated with VP shunt inser-
Prevention options for VP shunt infections organisms seen on CSF Gram stain, or organisms found in blood culture. Excluded from the study were patients with ventriculoatrial shunt, lumboperitoneal shunt, pre-existent CSF infection, incomplete infection and microbiological data, and patients whose shunt was removed before a diagnosis of infection was determined. CSF was obtained through needle aspiration of the shunt reservoir.
Table 1. Demographic and epidemiological characteristics of patients with VP shunt insertion
Age (median and range) Gender Male Female Reason for shunt insertion Head trauma Brain tumor Intracranial hemorrahage Other Length of hospital stay *
VP infection (%) Non-infection (%) (n = 28) (n = 356) 38 (6-76) 43 (5-82) 25 (89)* 3 (11)
220 (62) 136 (38)
11 (39) 9 (32) 5 (18) 3 (11) 47 (6-148)*
51 (14) 165 (46) 38 (11) 102 (29) 22 (5-271)
P < 0.05.
tion and to assess what factors, if any, were the essential steps in the prevention of VP shunt infection. Patient and method A retrospective chart review was performed of all patients aged 2 years or older submitted to VP shunt surgery between January 2006 and July 2010 at the Neurosurgery Department of Huashan Hospital, Fudan University, which is the largest neurosurgical medical center in Shanghai, China, where, collectively, approximately 8,000 neurosurgical procedures are performed annually. Clinical and laboratory data, the clinical condition requiring shunt procedure, as well as information on response to treatment and outcome were obtained from each patient’s medical records. As preoperative antibiotic prophylaxis, a single dose of vancomycin was routinely used before 2007. It generally consisted of the third-generation cephalosporin and vancomycin administered prior to skin incision from 2008. To define the shunt infected group, the criteria used in this study were established according to previous studies [9, 10]. VP shunt infection was defined as at least one of the following criteria was present: (1) isolation of the organism from 1 or more cerebrospinal fluid (CSF) cultures or (2) fever (temperature > 38°C), headache, neck stiffness, cranial nerve signs, or irritability without another recognized cause; a typical CSF findings such as: CSF leukocyte count > 250×106 cells/L, CSF total protein > 450 mg/L, CSF glucose level < 2.5 mmol/L,
19776
Statistical analysis Statistical analysis was performed using the SPSS 10.0 software package (Version 11.5; SPSS, Chicago, IL, USA). The variables are reported as absolute and relative frequencies and as the mean and standard deviation. Quantitative variables were analyzed with the Student t test or the Mann-Whitney test when appropriate. Qualitative variables were analyzed with the chi-square test with the Yates correction or the Fisher exact test (2-tailed) when necessary. A P value less than .05 was considered statistically significant. Results During the study period, 28 VP shunt infections resulted from 384 shunt procedures, giving an overall infection rate of 7.3%. The infection rate of VP shunt was 12.5% (12/96) and 10% (12/97) in 2006 and 2007, which significantly dropped to 2% and stabilized at lower level from 2008. The 28 patients with VP shunt infections including 3 pediatric patients and 25 adult patients (mean age: 38 years old) (Table 1). 89% of infections (25 episodes) involved male patients. The duration of symptoms before diagnosis occurred within 6 months of the procedure, ranging from 1 day to 158 days (mean, 18.8 days). Of the 28 infections, up to two-thirds (85%), the shunt infection occurred within 30 days post-operation, which was either the primary shunt insertions or shunt revisions. Late infections occurred in 4 cases, about two months, three months and five months after shunt revision, respectively. The clinical features of shunt infection were similar in children and adult (Table 2). Nearly
Int J Clin Exp Med 2015;8(10):19775-19780
Prevention options for VP shunt infections Table 2. Clinical features of VP shunt infection of 28 patients Parameters Clinical symptom Temperature > 38°C Consciousness disturbances Neck stiffness Headache Infection signal by CT scan Associated infection Pneumonia Intra-abdominal infection Urinary tract infection Underlying diseases Diabetes Cardiovascular disease Tuberculosis Hypokalemia CSF Leucocytes > 250×106 Glucose level were < 2.5 mmol/L Protein > 450 mg/L Time between shunt insertion and diagnosis of infection < 30 days > 30 days
all patients (96%) presented with fever > 38.0°C. Although consciousness disturbances was noted in 71% of cases, only 9 (32%) of 28 cases was found neck stiffness, and 21% of cases complained of headache. The cranial CT scan showed signs of shunt infection in 3 (11%) of 28 patients. Pneumonia was the most frequent associated infection (39%), followed for intra-abdominal infection (18%), and urinary tract infection (14%). In addition, 9 (32%) patients had other severe underlying diseases such as diabetes, cardiovascular disease, tuberculosis or hypokalemia. CSF studies revealed leukocytosis, elevated total protein and decreased glucose levels. CSF leucocytes were more than 250×106 in 75% of shunt infections, and 21 of 28 (75%) CSF samples displayed neutrophil predominance. CSF total protein levels were abnormal in 57% (16 of 28), with a median of 982.48±743.91 mg/L (range, 106±3530 mg/d). CSF glucose level were < 2.5 mmol/L in 23 patients (82%), with a median of 1.98±0.86 mmol/L (range, 0.56-3.6 mmol/L). The pathogens causing shunt infection were determined in 26 episodes (93%, Table 3). In 19777
No. of patients (%) 27 (96) 20 (71) 9 (32) 6 (21) 3 (11) 11 (39) 5 (18) 4 (14) 4 (14) 3 (11) 1 (4) 1 (4)
two patients, no causative pathogen was identified. The most prevalent organisms were gram-negative rods episodes (15), S. aureus (4), S. epidermidis (3), Staphylococcus haemolyticus (2), coagulase-negative staphylococci (1), enterococcus faecium (1). Of the 15 gramnegative rods episodes, 5 (33%) were due to acinetobacter baumannii, 4 (27%) were due to klebsiella pneumoniae, 2 (13%) were due to Pseudomonas aeruginosa and Pseudomonas maltophilia, respectively, and one each by enterobacter cloacae and ent. agglomerans.
All the patients received appropriate antibiotic treatments. VP 21 (75) devices were promptly removed from the 15 patients. In 13 epi23 (82) sodes (46%), neurosurgical de16 (57) vices was not removed because of good response to antibiotic 24 (86) treatment alone (10 episodes), 4 (14) or the deterioration of the patient’s complications (3 episodes). When the diagnosis of gram-negative rods meningitis was established, carbopenem antibiotics (meropenem or imipenem) was used in 9 (64%) of 14 cases, with or without amikacin; the rest of cases were treated with a thirdgeneration cephalosporin and rifampin. Five patients with Staphylococcus aureus and coagulase-negative staphylococci infection were treated with vancomycin. In addition, 3 patients received intrathecal therapy with gentamicin and vancomycin. The response was successful in 23 cases. Overall mortality was 18% (5/28). Three of the five expired patients died in septic shock, whereas the remaining patient died from disseminated intravascular coagulation and multiple organ dysfunction syndromes. Discussion Shunt-associated infection is a severe complication of VP shunt therapy for hydrocephalus. Despite the appropriate care received by patients, VP shunt infection remains a major cause of morbidity and even mortality [11, 12]. An infection rate of 6.98% was observed in the present study. Multiple factors have been Int J Clin Exp Med 2015;8(10):19775-19780
Prevention options for VP shunt infections Table 3. Summary of clinical data in 28 patients with VP shunt infection Case 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28
Shunt removal No Yes Yes Yes No Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes No Yes No Yes Yes Yes Yes Yes Yes No Yes
Causative pathogen klebsiella pneumoniae klebsiella pneumoniae klebsiella pneumoniae S. epidermidis S. aureus Pseudomonas aeruginosa Pseudomonas maltophilia acinetobacter baumannii enterobacter cloacae S. aureus acinetobacter baumannii acinetobacter baumannii acinetobacter baumannii coagulase-negative staphylococci S. epidermidis staphylococcus haemolyticus S. aureus Pseudomonas maltophilia Pseudomonas aeruginosa negtive enterococcus faecium staphylococcus haemolyticus S. epidermidis ent.agglomerans klebsiella pneumoniae S. aureus negtive acinetobacter baumannii
brought up as predisposing to infection after VP shunt, in particular the age at shunt insertion, cause of hydrocephalus, the experience of the neurosurgeon, the cause of hydrocephalus, operative room traffic and perioperative antibiotics [4-6]. In accordance with other studies, the majority of shunt-associated infections in the present study occurred within 30 days after shunt insertions. Fever and consciousness disturbances were the primary clinical symptoms among the patients, whereas typical neurological signs-such as neck stiffness or headache-were present in less than one-half of the episodes. This phenomenon reveals the characteristics of implant-associated infections are typically caused by low-virulence and slowly replicating organisms growing on the shunt surface in biofilms [13]. Early detection of shunt infection is usually difficult because of its nonspecific clinical signs, once patients 19778
Antibody treatment (mg/kg/day) Imipenem Amikacin Meropenem Meropenem Vancomycin Teicoplanin Vancomycin Teicoplanin Rifampin Ceftazidime Imipenem Amikacin Meropenem Vancomycin Amikacin Fosfomycin Meropenem Ceftazidime Vancomycin Gentamicin Vancomycin Vancomycin Fosfomycin Vancomycin Teicoplanin Rifampin Imipenem Ceftazidime Gentamicin Vancomycin Amikacin Teicoplanin Imipenem Amikacin Meropenem Vancomycin Vancomycin Imipenem Amikacin
Intrathecal therapy No Yes No Yes No No No No No No No Yes No Yes No No No No No No Yes No No No No Yes No No
Outcomes Died Survival Survival Survival Died Survival Survival Survival Survival Survival Died Survival Survival Survival Survival Survival Survival Died Survival Survival Survival Died Survival Survival Survival Survival Survival Survival
with fever and delayed recovery of consciousness after a VP procedure, the occurrence of central nervous system infection should be suspected. The main causative agents of shunt infections are gram-negative rods episodes in our study. This is contrary to other studies, which have shown the most common episodes to be grampositive bacterium [4, 6, 7]. The infection rate of gram-negative bacterium was higher relatively in our study, it can be speculated that the routine use of antacids leads to bacterial colonization of the stomach with Gram-negative strains may figure out why a high percentage of the infection are caused by gram-negative rods in our study [14, 15]. In a recent study reported in our hospital the results are similar to ours that the most freInt J Clin Exp Med 2015;8(10):19775-19780
Prevention options for VP shunt infections quent isolated organism of nosocomial infections is Gram-negative microorganisms. This finding supports the contention that simultaneous infections in other parts of the body which were caused by the same gram negative microorganisms may be relevant to that higher incidence. S. epidermidis, S. aureus, and Staphylococcus haemolyticus were the most frequent isolate encountered in our gram-positive series. Some patients have received intrathecal therapy with amikacin and vancomycin. A successful outcome was observed in 3 cases receiving intrathecal therapy, but there were not enough cases to draw reliable conclusions. The goal of antibiotic prophylaxis in shunt surgery is to prevent the shunt hardware with organisms, with particular lay stress on Grampositive bacteria. Historically, the organisms most frequently causing infections of VP shunt are the Staphylococcus species. The causative episodes are coagulase-negative staphylococci in 80% of cases, and most of the organisms isolated are methicillin resistant Staphylococcus epidermidis (MRSE) [16-19]. Antibiotics appropriate for prophylaxis is based on the above-mentioned literature, vancomycin was as used routinely antibody prophylaxis theory. However, the infection rate of VP shunt was higher (12.4% and 10%) in 2006 and 2007. When administrated alone, intra-ventricular vancomycin was associated with the increasing Gram-positive bacteria shunt infection. Thus, clinicians consider the initial antibiotic treatment with both Norvancomycin and Ceftriaxone may provide the satisfactory prophylactic effect. After this change in protocol, only 7 infections occurred in the next year and 2 cases found after that. VP shunt infections was broken down into 2% and stabilized at lower level from 2008. Of note, short-term perioperative antibiotic prophylaxis may be of benefit in preventing shunt infections. Despite satisfactory sensitivity of antimicrobial therapy, an overall mortality of 18% was observed. In postoperative cases, the removal of the infected CSF shunt is considered one of the most important parts of therapy [4, 6, 8]. We found that CSF shunt removal was associated with a lower mortality rate in patients with VP shunt-related infections, and, therefore, the diagnosis of bacterial meningitis in patients with CSF shunts should remind the neurosurgeon to remove foreign body devices. 19779
Conclusion In conclusion, although gram positive organisms were the most frequently seen nosocomial VP bacterial infection in neurosurgical patients, our data show that infections with gram-negative bacteria form a significant episodes percentage in the shunt infection in our center. In the absence of controlled trials, the result of this study may suggest that management option of infected shunts, which consists of perioperative antibiotic prophylaxis (in our survey vancomycin and a third generation cephalosporin) plus removal of the infected shunt, seems to be essential for the survival of patients with VP shunt infection. Acknowledgements This work was supported by the National Clinical Key Subject, the National Natural Science Foundation of China (grant number 81072070, 81471241, 81171133, and 81271375) and the Natural Science Foundation of Shanghai (grant number 15ZR1405600). Disclosure of conflict of interest None. Address correspondence to: Dr. Tao Zhang, Department of Laboratory Medicine, Huashan Hospital, Fudan University, 12 Central Urumqi Road, Shanghai 200040, China. E-mail: [email protected]
References [1]
[2]
[3]
[4]
[5]
Pople IK, Bayston R, and Hayward RD. Infection of cerebrospinal fluid shunts in infants: a study of etiological factors. J Neurosurg 1992; 77: 29-36. Drake JM, Kestle JR, Milner R, Cinalli G, Boop F, Piatt J Jr, Haines S, Schiff SJ, Cochrane DD, Steinbok P, MacNeil N. Randomized trial of cerebrospinal fluid shunt valve design in pediatric hydrocephalus. Neurosurg 1998; 43: 294-303 Shunts. In Infections Associated with Indwelling Medical Devices. In: Bisno AL, Waldvogel FA, editors. Washington: American Society for Microbiology; 1994. pp. 91-109. Piedra MP, Ragel BT, Dogan A, Coppa ND and Delashaw JB. Timing of cranioplasty after decompressive craniectomy for ischemic or hemorrhagic stroke. J Neurosurg 2013; 118: 109114 Reddy GK, Bollam P, and Caldito G. Ventriculoperitoneal shunt surgery and the risk of
Int J Clin Exp Med 2015;8(10):19775-19780
Prevention options for VP shunt infections shunt infection in patients with hydrocephalus: long-term single institution experience. World Neurosurg 2012; 8: 155-163. [6] Dallacasa P, Dappozzo A, Galassi E, Sandri F, Cocchi G and Masi M. Cerebrospinal fluid shunt infections in infants. Childs Nerv Syst 1995; 11: 643-649. [7] Turgut M, Alabaz D, Erbey F, Kocabas E, Erman T, Alhan E and Aksaray N. Cerebrospinal fluid shunt infections in children. Pediatr Neurosurg 2005; 41: 131-136. [8] Fan-Havard P, and Nahata MC. Treatment and prevention of infections of cerebrospinal fluid shunts. Clin Pharm 1987; 6: 866-880. [9] Wang KW, Chang WN, Shih TY, Huang CR, Tsai NW, Chang CS, Chuang YC, Liliang PC, Su TM, Rau CS, Tsai YD, Cheng BC, Hung PL, Chang CJ and Lu CH. Infection of cerebrospinal fluid shunts: causative pathogens, clinical features, and outcomes. Jpn JInfect Dis 2004; 57: 4448. [10] Garner JS, Jarvis WR, Emori TG, Horan TC and Hughes JM. CDC definitions for nosocomial infections, 1988. Am J Infect Control 1988; 116: 128-140. [11] Tuan TJ, Thorell EA, Hamblett NM, Kestle JR, Rosenfeld M, and Simon TD. Treatment and microbiology of repeated cerebrospinal fluid shunt infections in children. Pediatr Infect Dis 2011; 30: 731-735. [12] Warf BC, Bhai S, Kulkarni AV and Mugamba J. Shunt survival after failed endoscopic treatment of hydrocephalus. J Neurosurg Pediatr 2012; 10: 463-470.
19780
[13] Conen A, Walti LN, Merlo A, Fluckiger U, Battegay M and Trampuz A. Characteristics and treatment outcome of cerebrospinal fluid shunt-associated infections in adults: a retrospective analysis over an 11-year period. Clin Infect Dis 2008; 47: 73-82. [14] Aubert G, Jacquemond G, Pozzetto B, Duthel R, Boylot D, Brunon J and Dorche G. Pharmacokinetic evidence of imipenem efficacy in the treatment of Klebsiella pneumoniae nosocomial meningitis. J Antimicrob Chemother 1991; 28: 316-317. [15] Vallée E, Joly-Guillou ML, and BergogneBerezin E. Comparative activity of imipenem, ceftazidime and cefotaxime against Acinetobacter calcoaceticus. Presse Med 1990; 19: 588-591. [16] Mullan E, Lucas C, Mackie S and Carachi R. Audit of ventriculoperitoneal shunt infections in paediatric patients, 2006-2013. Scott Med J 2014; 59: 198-203. [17] Demetriades AK, and Bassi S. Antibiotic resistant infections with antibiotic-impregnated Bactiseal catheters for ventriculoperitoneal shunts. Br J Neurosurg 2011; 25: 671-673. [18] Bayston R, Ullas G, and Ashraf W. Action of linezolid or vancomycin on biofilms in ventriculoperitoneal shunts in vitro. Antimicrob Agents Chemother 2012; 6: 2842-2845. [19] Tomio R, Akiyama T, Shibao S and Yoshida K. Procalcitonin as an early diagnostic marker for ventriculoperitoneal shunt infections. Surg Infect (Larchmt) 2013; 14: 433-436.
Int J Clin Exp Med 2015;8(10):19775-19780
