Staphylococcus, first germ which was discovered the effectiveness of penicillin, remains a problem in germ respiratory infections. If his place is modest in community-acquired pneumonia, it is in the foreground in nosocomial infections by respiratory frequency and its prognosis. Therapeutically, the problems are multiple and related, its acquired resistance to antibiotics, knowing that Staphylococcus aureus (TO) methicillin resistant (SARM) is the most frequently encountered multiresistant bacteria in hospitals.
That is why we will try to answer two questions : In the probabilistic treatment of nosocomial pneumonia, what is the role of anti-staphylococcal antibiotics (whose spectrum is quite specific and limited to Gram-positive cocci for active molecules on methicillin-resistant strains) ? What is the optimal treatment of invasive multiresistant strains whose prognosis, worse than the infections methicillin susceptible strains, could be related to terrain or therapeutic care.
Place staphylococcus : epidemiological data
Epidemiological data confirm the importance of staphylococcus in all nosocomial infections. In the report of the Paris-Nord CCLIN 2003, the incidence of infections S. aureus is 0,82 for 1 000 hospital days and reached 0,98/1 000 days of hospitalization in short-stay services, is 0,48/100 admissions. In intensive care units and intensive care units, the prevalence of nosocomial infections is much higher and reaches 21 % in which EPIC 30 % are due to SA, first causative. Concerning nosocomial pneumonia, SA is also the first germ found : 24 % KISS study in Germany and 20 % in the USA (national nosocomial infection surveillance).
The frequency of resistant Staphylococcus aureus (SARM) may vary depending on the service, the hospital, the country : she reached 40 % Nosocomial infections SA in France in the study of CCLIN Paris-Nord, 60 % Nosocomial infections in intensive care in the European EPIC study, but only 14 % Nosocomial infections in intensive care in Germany KISS study. This frequency also depends on whether the nosocomial pneumonia is early (occurring in the first days of hospitalization) with a predominance of strains susceptible to methicillin, or late (after 3 or 4 days of hospitalization) with a predominance of strains resistant to methicillin. In the latter cases (late pneumonia), the other most common germ is Pseudomonas aeruginosa.
Staphylococcus aureus is a commensal bacterium of the skin and mucous membranes isolated in nearly 30 % patients and also present among caregivers, especially present in wetlands (nasal mucosa, throat and perineum), but can survive on inanimate surfaces (clothing, objects, instruments…), resulting in easy transmission. Pathogenicity is related to his character pyogenic, originally suppurative infections, but his blood borne favored by local microthromboses phenomena and production of toxins causing toxin syndromes. Among them, Leukocidin Panton Valentine is particularly daunting because the cause of necrotizing pneumonia and nosocomial transmission of producing this leucocidin strains has been reported.
The pathophysiology of pneumonia involves several possible mechanisms : contamination of airways by airborne route is practically not involved in this germ ; transmission by contiguity is rare ; transmission through blood from a remote focus (cutaneous, venous, endocarditis right heart ...) is possible and quite characteristic of staphylococcal infections (pulmonary drug addicts subjects such) ; but the main mechanism is linked to micro-aspiration. Several steps are required for the acquisition of pneumonia.
Colonization oropharyngeal :
Inhalations are from germs porting oropharyngeal and therefore either the seeds of the commensal flora worn by the patient prior to admission, or germs from the hospital environment. Thus in the study of Campbell et al, the frequency of pneumonia in SA is higher in SA patients at the nostrils admission (15,3 %) than in non-patients (2,1 %) SA and isolated strains of pneumonia are often identical to strains isolated from nasal cultures for admission.
Colonization tracheobronchial :
These micro-aspiration are the cause of tracheal colonization precedes nosocomial pneumonia. Thus in the study of George et al, tracheal colonization appears after or concurrently in the upper airway and above average 3 days the onset of pneumonia SA. The prevalence of bronchial colonization SA is evaluated in the study Allaouchiche et al. at 22 % sure 164 bronchoalveolar washes (dont 18 % with concentrations > 104 UFC/ml).
The tracheobronchial contamination (and thus secondarily colonization) could also be done directly from the environment to the lower airways, via endobronchial gestures (aspirations endotrachéales, endoscopies). Thus in the study of Pujol et al., Endoscopy is an independent risk factor for nosocomial pneumonia MRSA. However, endoscopy could intervene by promoting the introduction of a strong MRSA inoculum already present on the inner walls of the tube endotracheal all these patients often have a nasopharyngeal colonization prior or concomitant MRSA.
or parenchymal lung infection :
This colonization leads to infection depending on the size of the inoculum (volume and repeating the inhalations and higher pullulation). So on 220 mechanically ventilated patients and carriers of SA in bronchial secretions, 53 % were considered simply colonized, 26 % had tracheobronchitis and 19 % acute pneumonia. Eighty percent of patients who develop nosocomial pneumonia SA previously had nasal colonization to SA.
Risk factors :
Besides the risk factors for nosocomial infections common to all organisms whose pathophysiological mechanism essentially within micro-aspiration, there seem to be risk factors for MRSA or SA.
Infection Risk Factors in SA :
- Nasal colonization SA is associated with a higher risk of pneumonia SA.
- If age appears to be a risk factor for pneumonia in community SA, this is the presence of multiple co-morbidities (on average 3) which is a risk factor for nosocomial pneumonia SA : coma, cancer, corticosteroids, surgery, BPCO, diabetes, skin lesions and the presence of invasive equipment.
- Coma and brain injury are particular risk factors for colonization and infection in SA. In the study by Rello et al., coma pneumonia is the only risk factor for nosocomial SA in multivariate analysis.
In that of Pujol et al., 55 % of patients with nosocomial pneumonia SA methicillin susceptible have a brain injury and / or coma. The frequency of pneumonia in SA is higher in patients with head trauma in several studies : 20 % versus 3,8 % and 25,9 % versus 1,2 %. SA is isolated from the lower respiratory tract more frequently in patients with head injury than other patients between 1 and 6 days after ICU admission. This increased frequency of colonization and infections in SA in neuro-surgical patients could be related to the management of these patients treated with hyperventilation, sedation, hypothermia and corticosteroids.
It is due to a specific alteration of the defenses of the respiratory tract of patients rather than increase exposure to SA strains. So, in the study of Inglis et al., immunotyping the isolated strains in SA 17 cranial trauma patients failed to find external source, or patient to patient transmission, and that of Pujol et al., the majority of SA strains were sensitive to methicillin. The particular susceptibility of these patients would explain the early onset of nosocomial pneumonia in SA in these patients and therefore the responsibility of stem carried by the patient on admission often sensitive to methicillin.
Infection Risk Factors MRSA :
The duration of hospitalization and intensive care unit length of stay are correlated with the occurrence of nosocomial pneumonia MRSA.
Antibiotic therapy may have a protective effect on the occurrence of early nosocomial pneumonia by reducing the colonization by Gram-negative bacilli (Haemophilus influenzae) and probably Streptococcus pneumoniae, but it increases the risk of selection of resistant organisms and late nosocomial pneumonia. Thus multiple prior antibiotic, the duration of antibiotic therapy are independent risk factors for nosocomial pneumonia MRSA. In the study by Weber et al., Fluoroquinolones are a factor associated with MRSA infection. Selection pressure by a mono-quinolone antibiotic with a, otherwise, been reported to cause nosocomial pneumonia in SA in a small series where there was a control group treated with dual therapy.
Intubation and mechanical ventilation are risk factors : in the study of Pujol et al., 63 % of the 347 intubated patients had MRSA in the respiratory tract but only 6 % of the 33 non-ventilated patients (p < 0,001) and the length of intubation than 3 days appears as an independent risk factor for pulmonary MRSA. In the study of Trouillet et al., the upper ventilation 7 days is also a multivariate risk factor for infection resistant germs including MRSA.
Clinical and paraclinical : there he has signs suggestive of pneumonia in SA ?
The clinical and radiological picture seems somewhat specific in relation to that of nosocomial pneumonia caused by other bacteria : fever (86 % cases), expectorations purulentes (93 %) bilateral pneumonia (33 %) pleural effusion (19 %), empyème (4 %) in the study of Pujol et al.. Hyperbilirubinemia is often present and already reported in septicemia SA. Microbiologically, 8 % are associated with bacteremia, 4 % empyema and 38 % are polymicrobial.
Few elements also distinguish pneumonia MRSA and those to SAMS : no difference in clinical (fever, purulent secretions, existence of bacteraemia, bilateral radiographic, presence of pleural effusion or empyema). In one study, leukocytosis is a little higher in lung MRSA in nosocomial pneumonia early to SAMS. Associated germs are varied in early pneumonia to SAMS (pneumococcal, Haemophilus influenzae ou Streptococcus viridans), types of nosocomial (Pseudomonas, Acinetobacter…) in late pneumonia at SAMS or MRSA.
The clinical picture is not specific to SA and not to guide early treatment before microbiological results that are often available only secondarily. Useful items and early can still be made by microbiology : respiratory systematic microbiological samples in previous days have predictive value even knowledge of nasal carriage of SA by the patient. The finding of Gram-positive cocci in clusters under the microscope in bronchial samples used to quickly guide treatment.
The morbidity of nosocomial pneumonia in SA is important with, in particular, the extension length of stay in ICU and hospital. Mortality in these patients is high but the share attributable to nosocomial infection is difficult to assess.
Several studies report particularly high mortality rate for pneumonia in SA : 70 % in the study of Lentino et al. and up 84 % in bacteremic forms Watanakunakorn et al. The methicillin resistance is an aggravating factor in some studies (mortality times 3 for MRSA compared to the SAMS in the study Ibelings et al., but more studies are divergent. In the study by Pujol et al., Mortality is high in early nosocomial pneumonia in SAMS (43 %) and late in MRSA pneumonia (56 %), but lower in late nosocomial pneumonia at SAMS (30 %) : the poor prognosis of early pneumonia is probably related to a particular susceptibility of patients with a particular profile (head trauma, coma) as has been mentioned above. The prognostic difference between SAMS and MRSA pneumonia in late could be attributed either to a different patient profile (indices McCabe and SAPS higher in the MRSA group, or to a lower efficiency of vancomycin on MRSA that the ß-lactams of MSSA as suggested by the work of Bodi et al. In the study by Combes et al. covering only patients whose initial treatment is suitable, mortality is related in multivariate analysis to the field (age, number of organ dysfunctions) and not to the resistance to methicillin, suggesting that it would intervene in the forecast through the land on which it occurs and the more frequent inadequacy of initial treatment it engenders.
Antibiotic susceptibility :
The first S strains. aureus resistant to penicillin were identified 1942. The proportion of resistant strains has largely increased since ; today 90 % SA are resistant to penicillin G. In the early 1960, development antistaphylococcal penicillin family of methicillin was followed by the rapid emergence of methicillin-resistant SA. Since the early 1980, MRSA has become endemic in most hospitals worldwide.
The resistance to ß-lactam SA falls 2 mechanisms : hydrolysis by penicillinase (having a low affinity for methicillin and itself hydrolysed by inhibitors of? -lactamase thereby restoring sensitivity to aminopenicillins) and PLP changes. What 2e mécanisme induit une résistance du SA in the méticilline and the ensemble des β-lactamines. This resistance to β-lactam antibiotics is most often associated with resistance to fluoroquinolones. However, these strains remain susceptible to a number of molecules acting independently of PLP : glycopeptides, les synergistines, rifampicin, fusidic acid, fosfomycin and some aminoglycosides.
The evolution of MRSA resistance to aminoglycosides is special because we went over 90 % resistance to all aminoglycosides to sensitivity over 80 % for gentamicin. However this possibility of use of aminoglycosides should be discussed in the light of enhanced nephrotoxicity of glycopeptide Association - aminoglycoside. Rifampicin is certainly the most active antibiotic against SA quickly ; but it is also the antibiotic that selects the most resistant mutants requiring an actual bacteriological association and pharmacological (that is to say the presence in adequate concentrations and at the same time to the infectious site 2 molecules). It is the same with fosfomycin is, in addition, responsible for a significant sodium intake in frail patients.
In the years 1990, appeared strains of SA decreased susceptibility to vancomycin (AS). MIC range 8 and 16 mg / ml defines a reduced susceptibility of strain or intermediate, and an MIC greater than 32 mg / ml defines a strain resistant to vancomycin. in addition 1996, the first documentation has been reported in Japanese children. Since, number of similar cases have been documented in hospitals from different countries (United States, France). Factors associated with the emergence of these GISA are prolonged treatment glycopeptide (between 4 and 18 weeks) among patients with MRSA infections, the need for dialysis, exposure to invasive procedures and the presence of foreign bodies, and finally the proximity of a patient colonized by GISA.
Methods of treatment :
Systemic infections MSSA should be treated with parenteral antibiotics bactericidal, obtained by combining a penicillin group M with an aminoglycoside or a fluoroquinolone.
Antibiotic treatment of MRSA infections based on glycopeptides (vancomycin and teicoplanin). It is better to control serum levels of these products due to inter- and intra-individual variation, the risk of nephrotoxicity and risk of inefficiency (if too low valley). Recommended rates are 40 mg / l at the peak, and 20 mg / l in the valley. Continuous administration is also proposed. Bactericidal activity is slow to get, glycopeptides must be prescribed in combination in severe infections, avoiding frequently inactive quinolones on MRSA. The association aminoglycosides (including gentamicin) is the most common.
Other combinations are possible with the fusidic acid, la fosfomycine, rifampicin and synergistines, if necessary by shifting the administration of rifampicin or fosfomycin to avoid a situation in monotherapy 12 at 48 first hours would be responsible for the appearance of mutants resistant to these 2 antibiotics. The optimal treatment duration is unknown ; it is conventionally 15 days but a shorter period is proposed. Thus in the study of Chastre et al, a period of 8 days has the same efficacy in MRSA pneumonia in patients whose initial antibiotic therapy were appropriate.
Among the recent active molecules on SA, quinupristin / dalfopristin and linezolid showed equivalent efficacy with vancomycin in the treatment of nosocomial pneumonia. However, it is noted, in these studies, enrollment subgroups SA and MRSA limited and probably insufficient to show a difference. In the overall population, a high proportion of patients infected with other germs and receiving aztreonam in association with molecules to study mask any differences. These molecules thus appear as alternatives to glycopeptides which we can not ignore some drawbacks : potential toxicity (including kidney) and poor pharmacokinetics Realizing the need for assays for dose adjustment and probably a limitation of product efficacy.
If glycopeptides remain the reference molecules in the treatment of MRSA infections, However, it makes sense to ask the question of preferential use of a molecule as linezolid whose theoretical advantages are many : better tolerance, better lung pharmacokinetic, can administration by IV and oral and absence of cross resistance with other antistaphylococcal. Note however that the cost remains high and uncertainty about the possibility of emergence of resistant strains but still rare in this day.
Therapeutic indications :
Initially, the treatment of nosocomial pneumonia is, most of the time, empirical, or the causative organism is not known(s) before the result of withdrawals. The inadequacy of this first antibiotic treatment strongly influences the prognosis resulting in significant mortality and calls for greater efforts to reduce the risk of inappropriate treatment. It will take into account the specific risk factors on clinical and epidemiological data and the results of a microscopic examination of respiratory secretions It may target, According to the case, Community kind of germs (in case of early nosocomial pneumonia) or, more often, potentially multiresistant bacteria, including Gram-negative bacilli and possibly MRSA.
A ß-lactam combined with an aminoglycoside or a fluoroquinolone is an option. This treatment "broad spectrum", but not target SA, may nevertheless be effective on sensitive SA methicillin. The choice, first-line, treatment for MRSA, rarely alone, most often associated with a wider treatment, must be in the presence of known risk factors, a particular local epidemiology or severity of signs involving rapidly life-threatening. Considering, Firstly, the importance of the initial choice of antibiotic therapy in the prognosis of nosocomial pneumonia noted in numerous studies and, on the other hand, the importance of ITS in the epidemiology of nosocomial pneumonia, it is legitimate to question whether to include MRSA systematically in the initial treatment spectrum. This attitude, however, does not seem desirable as it presents several risks : increased iatrogenic risk, lesser efficacy on MSSA, impact on the evolution of resistance, including the proliferation of GISA strains.,
SA acquired a prominent place in nosocomial pneumonia in frequency and severity and poses unresolved therapeutic problems due, essentially, its resistance to antibiotics (resistance to methicillin and frequent real, resistance to glycopeptides threatening for years to come) and the difficulties of managing major antistaphylococcal. The desire to have an initial empirical treatment effective immediately, must discuss the addition of a molecule effective on MRSA rest of broad-spectrum antibiotics, even if there are major drawbacks of this attitude : toxicity, cost, ecological risk (evolution of resistance).
It is therefore important to recognize situations at risk of MRSA infection justifying such treatment. The appearance of new molecules antistaphylococcal should improve the management of these patients even if their place (next glycopeptides) is not yet definitively established. All these therapeutic issues related to antibiotic resistance of SA, highlights the importance of measures to limit the transmission of these strains resistant to first and foremost involved hand washing and the broad use of alcoholic solutions.