|dc.description.abstract||Lyme disease is the most common human tick borne disease in the northern hemisphere, and
the southern coastline of Norway is a high endemic region regarding Lyme disease.
European Lyme neuroborreliosis (LNB) often presents with a sub acute painful lymphocytic
meningoradiculitis (Bannwarth syndrome) with or without paresis in the abdominal wall, the
limbs or muscles innervated by cranial nerves. Central nervous system involvement as
encephalitis, myelitis and vasculitis is rare.
Most LNB patients experience marked improvement in neurological symptoms within weeks
to a few months after antibiotic treatment, but years after treatment 10 – 50% report persisting
or new symptoms including fatigue, concentration difficulties and musculoskeletal problems.
Remaining complaints after adequately treated Bb infections are often named Post Lyme
Disease Syndrome (PLDS). The prevalence and impact of PLDS is debated since similar
symptoms are common in the general population, and there are few European controlled
studies on the issue. Most studies on outcome after LNB are conducted in the US, and as
Borrelia genotype and the clinical picture of Lyme disease in the US differ somewhat from
what we find in Europe, the study results are not necessarily transferable to European patients.
Our aim was to assess the long-term impact of LNB on Health Related Quality of Life
(HRQoL) in a controlled study of well-characterized adult European LNB patients.
We also wanted to compare the neuropsychological (NP) functioning by assessing
executive/attention functions, processing speed and memory in a group of adult LNB patients
30 months after treatment to a matched control group. Finally, we wanted to identify clinical,
demographical or laboratory factors associated with a reduced HRQoL and fatigue after
treatment of LNB.
Patients and methods
A cohort of 50 patients was followed for 30 months after treatment for LNB. The patients
were recruited from a treatment study conducted in southern Norway comparing per oral
doxycycline to intravenous ceftriaxone. All patients were living in the geographical region of
Agder Counties, and received treatment between May 2004 and December 2007.
The LNB patients brought a control person from the same geographical area, matched for age,
gender and education level. Exclusion criterion for the controls was a history of
acknowledged LNB. At basis pre-treatment and at 4, 12 and 30 months we did a semi structured interview, a
clinical score, and spinal and blood tap. At 30 months NP functioning was assessed, and all
participants completed the following questionnaires: the Fatigue Severity Scale (FSS), the
Montgomery-Åsberg-Depression Rating Scale (MADRS), and the Short Form-36 (SF-36: A
HRQoL questionnaire including the sum scores Mental Component Summary (MCS) and
Physical Component Summary (PCS)). They were asked about previous and current
coexisting diseases, psychological distress and subjective complaints. A composite clinical
score made for the treatment study, summarizing subjective complaints and objective
findings, was used to assess clinical status. Non-complete recovery was defined as more than
1 point score on the composite clinical score.
The four NP tests in our test panel consisted of 23 subtasks, and we calculated a sumscore
expressing the number of NP subtasks with the scores ≤-1 SD from the mean in the control group
(range 0-23). The sumscores were then categorized into three groups: Normal: 1-5(≤-1 SD
from the mean sumscore in the control group), deficit: 6-8 (>-1≤2 SD from the mean
sumscore in the control group) and impairment: 9-23 (>-2 SD from the mean sumscore in the
In the study regarding risk factors we did a univariate analysis comparing independent
demographical, clinical and laboratory data to the PCS, MCS and FSS scores to look for
associations. The variables which were associated with the outcome in the univariate analyses
were analyzed further in a multiple regression model.
Before treatment 80 % of the patients had a complete or partial Bannwart syndrome, and 8 %
had symptoms suggesting involvement of the central nervous system (myelitis, ataxia and
confusion). Fifty percent were treated with oral doxycycline and 50 % with IV ceftriaxone.
Sixty-eight percent were classified as definite LNB according to the criteria of the European
Foundation of Neurological Society, and 32 % as possible LNB. Mean age at follow up was
55, and 58 % were male.
Paper I: LNB treated patients had reduced HRQoL compared to controls as assessed with the
SF-36 summary components PCS (P<0.001) and MCS ( p=0.010) 30 months after
treatment. The patients scored lower on all the eight subscales of the SF-36, except for bodily
pain. The LNB patients who reported complete recovery (56%) had similar HRQoL scores as
the controls. Paper II: LNB treated patients scored lower on four NP subtasks assessing
executive/attention functions, processing speed, visual and verbal memory, as compared to
matched controls: Stroop test 4 (P=0.015), TMT 5 (P=0.004) Digit Symbol recall (P=0.038)
and CVLT list B (P=0.003). The distribution of global NP function indicates that most of the
LNB treated patients perform comparable to controls, while a small subgroup have a
debilitating long-term course with cognitive impairment. Fatigue, depression, neurological
deficits or HRQoL at 30 months after treatment were not associated with the global NP sum
score. Eighteen out of 50 patients (36%) had objective findings in terms of neurological
deficits and/or cognitive impairment.
Paper III: Delayed start of treatment and remaining complaints 4 and 12 months after
treatment seem to predict a worse outcome with respect to HRQoL. Delayed start f treatment,
a more severe disease pre-treatment and remaining complains at 4 and 12 months after
treatment seem to predict more fatigue at 30 months. Age, gender, educational level,
diagnostic accuracy, treatment option, signs of infection in the central nervous pre-treatment
or coexisting somatic diseases or psychological distress were not associated with HRQoL
outcome 30 months after treatment in our cohort, neither were any of the assessed CSF
findings before treatment or during follow-up .
1. HRQoL was reduced in well-characterized European patients treated for LNB with a
current recommended antibiotic regimen 30 months earlier, as compared to matched controls.
The LNB treated patients were not more depressed and did not report more pain than the
controls. Fatigue was the most disturbing persisting complaint, and was negatively associated
with HRQoL. Mild neurological deficits were found in 28 % of the patients, and seemed to
influence negative on the physical HRQoL and fatigue scores. The patients who reported
subjective recovery had the same HRQoL as the controls.
2. Most of the patients who were treated for European LNB 30 months earlier had comparable
NP functioning to matched controls, but a small subgroup had cognitive impairments
regarding attention/ excecutive function, processing speed and memory that could affect their
daily life. The LNB treated patients with complete recovery had similar NP functioning as the
controls. We did not find any association between NP test results and HRQoL or fatigue.
3. It seems as a more serious LNB disease and a longer duration of symptoms before
treatment can reduce HRQoL, and as symptom duration more than 6 weeks before treatment, a more severe disease and non-complete recovery at four and 12 months predict a higher
burden of fatigue 30 months after treatment. We did not find that any laboratory data
predicted outcome after treated LNB, or that any CSF finding indicated an active Bb
infection 30 months after treatment. Gender, age, comorbidity, signs of pre-treatment
infection of the central nervous system or CSF findings before and during follow-up were not
associated with HRQoL or fatigue at 30 months.
4. Thirty months after treatment of LNB 18 out of 50 patients (36%) had objective findings in
terms of neurological deficits and/or cognitive impairment.||en