Predictive scores failing at identifying psychiatric disabilities following childhood bacterial meningitis calls for revision of current follow-up guidelines

Figures & data

Figure 1. Systematic search and stepwise review.

This figure shows the results of the systematic PubMed search that we conducted to identify predictive scores for our study, and the stepwise process of reviewing these publications. Number of publications for each step is shown in boxes.

¹At this step, reference lists of all 24 publications were reviewed to identify any additional relevant publications. This resulted in an additional 19 publications that were also read in full. None of these were deemed relevant.

Table 1. Criteria of the five predictive scores.

At admission	Aronin [16]	Herson-Todd [17]	MeningiSSS [14]	Niklasson [16]	Simple Luanda Scale [14]
Circulation	SBP ≤ 90 mmHg or > 40 mmHg decrease (1p)^a	SBP < 60 mmHg (1p)^a	Peripheral failure (1p) Circulatory shock (2p)^b	SBP ≤ 100 mmHg (1p)^a	—
Respiration	—	—	Moderate dyspnoea (1p)^c	—	Moderate dyspnoea (1p)
					Severe dyspnoea (2p)^c
Mental status	Altered mental status (1p)	Coma (3p)	Reduced consciousness (2p)	—	Reduced consciousness (5p)
					Coma (10p)
Body temperature	—	<36.6 °C (2p)	—	> 40 °C (1p)	—
Cerebrospinal fluid findings	—	WBC < 1,000 cells/µL (1p)	WBC < 1,000 cells/µL (1p)	WBC < 100 cells/µL (1p)	—
		Glucose < 1.1 mmol/L (0.5p)	Glucose ≤ 0.6mmol/L (1p)
			Protein ≥ 2,500 mg/L (1p)
Other laboratory findings	—	Haemoglobin < 110 g/L (1p)	Low WBC (2p)^d	WBC < 15 × 10^9/L (1p)	—
				Platelets < 100 × 10^9/L (1p)
Duration of symptoms at admission	—	Symptoms > 3 days (0.5p)	Fever > 7 days (1p)	Petechiae < 12 hours (1p)	Symptoms 4–7 days (1p)
					Symptoms > 8 days (2p)
Other criteria	—	Age < 12 months (1p)	—	—	No electricity at home (2p)
Any time
Seizures	Any (1p)	Any (2p)	Any (2p)	—	Focal (1p)
					General (2p)
Cut-off level	≥ 2 points	≥ 4.5 points	≥ 6 points	≥ 3 points	≥ 7 points

This table shows the scoring criteria for the predictive scores tested in our study. These criteria were assessed based on clinical and laboratory findings at admission to the hospital including the first 24 h thereafter, except for level of consciousness and seizures that were considered positive if they occurred at any point during the hospital stay. The cut-off level indicates high risk of adverse outcomes.

CSF: Cerebrospinal fluid; WBC: White blood cells.

^aIf blood pressure at admission was missing, the criteria was deemed fulfilled if the patient needed inotropic drugs to maintain circulation.

^bImpact on circulation was defined and graded according to the age correlated systemic inflammatory response syndrome (SIRS) criteria, either as peripheral circulatory failure defined as tachycardia or clinical signs of peripheral circulatory failure, or as circulatory chock.

^cImpact on respiration was defined and graded according to the age correlated systemic inflammatory response syndrome (SIRS) criteria, either as moderate dyspnoea defined as tachypnoea or as severe dyspnoea defined as need of respirator.

^dImpact on white blood cell count in blood was defined according to the age correlated systemic inflammatory response syndrome (SIRS) criteria.

Table 2. General features of the 73 included patients including permanent psychiatric disabilities.

Patients’ characteristics and causative pathogen	No.	Occurrence /means
Age at admission, mean ± SD (years:months)	—	3:9 ± 5:0
Sex distribution (%)	—	—-
Girls	37	51
Boys	36	49
Duration of illness before admission, mean ± SD (days)	—	1.8 ± 1.5
Causative pathogen	—	—
- H. influenzae (%)	38	52
- S. pneumoniae (%)	18	25
- N. meningitidis (%)	7	10
- Other identified pathogens (%)^a	4	5
- Unknown pathogen (%)	6	8
In-hospital characteristics	No.	Occurrence /means
Duration of antibiotic treatment, mean ± SD (days)	—	12 ± 8
Duration of hospital stay, mean ± SD (days)	—	14 ± 6
Complications	—	—
- Intracerebral structural injury (%)^b	7	10
- Pathological neurological findings (%)	14	19
- Repeated or prolonged seizures (%)	13	18
- Other complications (%)^b	18	25
Cerebral function at discharge^d	No.	Occurrence
No disabilities (PCPC category one) (%)	60	82
Disabilities (%)	—	—-
Mild (PCPC category two)	11	15
Moderate (PCPC category three)	2	3
Severe (PCPC category four)	0	0
Disabilities	No.	Occurrence
Psychiatric disease (%)	22	30
Concentration or learning difficulties (%)	15	21
Need of special education (%)	6	8
Hearing impairments (%)	22	30
Neurological deficits (%)	17	23
Epilepsy (%)	8	11

This table shows the general features for the 73 patients included in our study, treatment strategies during the hospital stay, short-term outcome and occurrence of disabilities noted during the observation period. All 73 patients were included in all analyses. In the table, the number of affected patients (No.) are presented in the first column, followed by the corresponding occurrence presented in the second column as percentages or as means with standard deviation (SD).

^aThis category included three cases of group B streptococci and one case each of group A streptococci and Escherichia coli.

^bThis category included six cases of aseptic hygromas, and one cases of ischaemic injury. All cases of intracerebral structural injury were diagnosed via brain computed tomography except in one case of aseptic hygroma diagnosed via repeated brain ultrasounds.

^cThis category included four cases of reactive arthritis, four cases of coagulation problems, three cases of respiratory problems, two cases of herpes stomatitis or other peripheral viral infections, two cases of SIADH, one case of kidney failure, one case of osteitis, one cases of skin necrosis, one case of vision impairment, one case of septic arthritis, and one case of hydrocele.

^dCerebral function at discharge was retrospectively graded using the Paediatric Cerebral Performance Category Scale (PCPC) [15] where a normal function was defined as PCPC category 1 and disabilities were defined as PCPC categories 2–5. In five patients, these disabilities were previously known and were due to pre-existing diseases.

Figure 2. Length of the observational period.

This figure shows the length of the observational period for the 73 included patients using a Kaplan-Meier Curve. For each year, the percentage of patients having an observational period of at least this number of years is indicated using a vertical bar.

Table 3. Predictive scores’ ability to predict psychiatric disabilities.

	No.	AUC	Grade	Sens.	Spec.	PPV	NPV	p^Cut-Off
Psychiatric disease
Aronin	71	0.73	Fair	38	80	44	76	0.11
Herson-Todd	68	0.64	Poor	20	94	57	74	0.09
MeningiSSS	60	0.59	Failed	38	77	38	77	0.25
Niklasson	66	0.63	Poor	6	92	20	74	0.76
Simple Luanda Scale	71	0.67	Poor	33	82	44	75	0.16
Concentration or learning difficulties
Aronin	71	0.72	Fair	57	83	44	89	<0.01
Herson-Todd	68	0.70	Fair	15	91	29	82	0.50
MeningiSSS	60	0.59	Failed	40	76	25	86	0.30
Niklasson	66	0.56	Failed	8	93	20	80	0.99
Simple Luanda Scale	71	0.53	Failed	36	81	31	84	0.19
Need of special education
Aronin	71	0.92	Excellent	60	77	17	96	0.07
Herson-Todd	68	0.63	Poor	0	89	0	93	0.49
MeningiSSS	60	0.75	Fair	67	75	13	98	0.11
Niklasson	66	0.56	Failed	0	92	0	93	0.56
Simple Luanda Scale	71	0.62	Poor	40	79	13	95	0.33

This table shows the ability of the five predictive scores at identifying children with bacterial meningitis later developing psychiatric disease, concentration or learning difficulties, or any type of psychiatric disability. First, the number of patients included for each predictive score (No.) are shown, followed by the area under the curve (AUC) obtained from the receiver operating curve characteristics analysis together with the performance category grading (Grade) based on this result [19]. Lastly, each predictive score’s sensitivity (Sens.), specificity (Spec.), positive predictive value (PPV) and negative predictive value (NPV) when tested at their cut-off levels using a chi-square test are shown, including the p-value of this analysis (p^Cut-Off).

^aThis category comprised of children with either psychiatric disease or with concentration or learning difficulties.

Bijlsma MW, Brouwer MC, Bossuyt PM, et al. Risk scores for outcome in bacterial meningitis: systematic review and external validation study. J Infect. 2016;73(5):393–401.

 PubMed Web of Science ®Google Scholar

Olson D, Lamb MM, Gaensbauer JT, et al. Risk factors for death and major morbidity in guatemalan children with acute bacterial meningitis. Pediatr Infect Dis J. 2015;34(7):724–728.

 PubMed Web of Science ®Google Scholar

Johansson Kostenniemi U, Karlsson L, Silfverdal SA, et al. MeningiSSS: a new predictive score to support decision on invasive procedures to monitor or manage the intracerebral pressure in children with bacterial meningitis. Neurocrit Care. 2020;32(2):586–595.

 PubMed Web of Science ®Google Scholar

Volakli EMD, Sdougka MMD, Mantzafleri PE, et al. Functional outcome following pediatric intensive care: pediatric cerebral performance category (PCPC) and pediatric overall performance category (POPC) during a prospective two years follow-up period. Greek e J Perioper Med. 2015; 13:2–15.

 Google Scholar

Kleinbaum DG, Klein M. Logistic regression, a self-learning text, 3rd ed. New York: Springer; 2010.

 Google Scholar