##############################
P KZN CRP PN
Serum C‐reactive protein is a useful tool for prediction of complicated course in children's pneumonia
C‐reactive protein (CRP) is an acute‐phase protein synthesised by hepatocytes in the liver. CRP was found in the 1930's in sera of patients with pneumonia, since this protein precipitated with a soluble extract of Streptococcus pneumoniae, later identified as a C‐polysaccharide of the bacterial cell wall that is common to all pneumococcal strains.1 CRP belongs to the highly conserved pentraxin proteins, and in fact, CRP is the firstly found pattern recognising protein of innate immunity. From the 1940's onwards, the research on CRP was moved from the structure and production of the molecule to the identification of diseases in which CRP can rise. A Finnish study demonstrated as early as in 1959 that serum CRP rose in purulent meningitis but not in aseptic meningitis.2 However, the value of serum CRP as a useful tool in the separation of bacterial from viral infections was questioned until the late 1970's.3 Bit by bit, the usefulness of CRP was demonstrated in meningitis, epiglottitis, pyelonephritis, purulent arthritis and septic infections of children with leukaemia, including the screening of complicated courses of these infections.4 A common feature for these infections is that bacterial diagnoses can be rather reliably confirmed by cultural methods.
The association between serum CRP concentration and the aetiology of lower airway infections, such as children's pneumonia with an exception of bacteremic pneumonia, has been less clear. The problem is important, since bacteremic pneumonias only compose a few per cent of all community‐acquired pneumonia cases. Sometimes, serum CRP seems to increase also in viral pneumonia and vice versa does not increase in all bacterial pneumonia cases. The obvious reason for this unexpected result lies in the difficulty to diagnose bacterial aetiology of pneumonia or other lower airway infections. In non‐bacteremic cases, bacterial involvement can be studied with antigen detection tests in serum, urine or respiratory samples, with antibody or immune complex assays in acute or paired sera, and with staining, antigen detection or polymerase chain reaction in buffy coats of white blood cells.5 However, we do not know how common false‐negative and false‐positive results are in these tests, and none of them have pointed out to be sufficiently sensitive and specific for clinical practice in the diagnosis of bacterial pneumonia in children.6 Irrespective of the method used, samples from upper airways are not appropriate for studies on bacterial aetiology of pneumonia. Maybe, the principal problem has not been in the results obtained from serum CRP determinations but in the results obtained from microbiological examinations in non‐bacteremic pneumonia cases.
Many studies have documented that mixed viral and bacterial aetiologies, as well as mixed mycoplasmal and other bacterial aetiology, are common in children's pneumonia and other lower airway infections.7 Viruses pave way for secondary bacterial infections. If serum CRP rises in viral lower airway infection, the reason for that rise most often is an emergence of bacterial co‐infection, or sometimes the worsening of virus‐induced tissue damage, or rarely another inflammatory complication. On the other hand, there are mild bacterial infections with no significant CRP rises, which recover spontaneously without antibiotics or other treatments. Currently, CRP is considered to be an unspecific acute‐phase protein that reflects the degree of inflammation and tissue damage in different infectious and non‐infectious diseases.
C‐reactive protein rises in acute events associated with marked inflammation or tissue damage rather rapidly, within some hours. In bacteremic infections, a significant increase is usually reached within 12‐48 h, which, however, is a too long time for waiting in severe septic or other life‐threatening infections. The maximum level is usually reached in the second or third day. Serum CRP of <20 mg/L is usually considered as normal. Mild infections, including viral, mycoplasmal and those bacterial infections that improve spontaneously, can raise serum CRP concentration to the level of 20‐50 mg/L. Bacterial infections that need treatment raise serum CRP usually to 60‐120 mg/L, and severe, bacteremic or septic infections to 120 mg/L or higher. These limits are only directive. In a meta‐analysis including eight studies and 1230 children with community‐acquired pneumonia,8 serum CRP concentrations exceeding 35‐60 mg/L were measured significantly more often if pneumonia was associated with bacterial infection than with viral infection alone. In our previous study, serum CRP over 60 mg/L separated cases with evidence of pneumococcal aetiology from those with proved pure viral aetiology with a sensitivity of 0.26 and specificity of 0.83.9 Lower CRP levels need to be used for considering of antibiotics in infants. In the neonatal period and in immune deficient children, all even small rises in serum CRP must be noted and may indicate starting of antibiotics. However, there are no limits that tell to the doctor that antibiotics should be started for ambulatory patients who have no underlying illness. The biologic half‐time of CRP is about 20 h, which means that the concentration falls rather rapidly when the infection improves, and correspondingly, a slow decrease or a new rise means the risk of complications.
Yuval Barak‐Corren et al publish in this issue of the journal their interesting findings from Israel on the prognostic value of serum CRP in 2.4‐month to 17.7‐year‐old children, who were admitted for pneumonia to the emergency department of an academic teaching hospital during 4 years.10 The study design was a retrospective analysis of patient records obtained from the electronic files of the hospital, which were also charted manually for clinically significant data. Originally, the number of admitted children was 2561, but those with the discharge diagnosis viral pneumonia and those in whom CRP was not measured were excluded. The median age of included 814 children was 3.2 years, and 57% were boys. The quartile limits of serum CRP taken on admission were 40, 100 and 200 mg/L, respectively. Overall, 38.8% of admitted children were hospitalised, 2.2% required admission to the intensive care unit, and 15.2% presented with pleural effusion of which 28% were drained. Serum CRP in the fourth quartile predicted significantly all these four non‐beneficial outcomes. Bacteremia was present in 31 (9.5%) of those 327 in whom blood culture was taken, and serum CRP over 260 mg/L on admission significantly predicted positive blood culture. Incorporating CRP in a multivariate prediction model including clinical prognostic factors, which have been significant in other studies, provided an area‐under‐curve (AUC) of 0.96 for the outcome of chest draining. For other outcomes, the AUC was 0.90 for bacteremia, 0.82 for pleural effusion, 0.75 for hospitalisation and 0.64 for admission to intensive care.
Age, gender, white blood cell count, infiltration in chest radiograph, presence of respiratory symptoms and duration of illness were prognostic factors included in the models. Without these clinical factors, the AUC varied from 0.55 to 0.85 depending on the outcome in question.
The authors concluded that CRP is a useful prognostic marker when evaluating a child with suspected bacterial pneumonia and helps in identifying those who need to be hospitalised and will benefit from close follow‐up during hospitalisation.10 Serum C‐reactive protein is a useful tool for rapid detection of the risk of non‐beneficial course and that of complications in children's pneumonia irrespective of the original aetiology of infection. Certainly, most non‐beneficial courses and complications are caused by primary or secondary bacterial infections.
#######################
WW2 KPNESE LEFT WNDED TO DTH
No comments:
Post a Comment