The aim of the study was to determine whether the use of combination therapy consisting of manual therapy and proprioceptive neuromuscular facilitation (PNF) is more effective than the use of the techniques of manual therapy, PNF method or traditional physiotherapy as single methods in the treatment of non-specific low back pain.

The study was designed as single-blinded Randomized Controlled Trial (RCT) and conducted on a group of 200 patients of Hospital in Parczew. The patients were randomly divided into four 50-person groups: A used manual therapy, B - PNF, C - manual therapy and PNF, and D -kinesiotherapy. Pain intensity was measured using VAS and Laitinen's questionnaire. Functional disability was assessed using Oswestry Disability Index (ODI) and Back Pain Functional Scale (BPFS).

. The hypotheses were:

1. The greatest reduction of pain is observed in the combined therapy group consisting of manual therapy and PNF method
2. The range of movement of the lumbar spine is the most improved in the combination therapy group consisting of manual therapy and PNF method.
3. Neurological symptoms assessed by neurodynamic tests are reduced to the greatest extent in the combined therapy group consisting of manual therapy and PNF method.
4. The degree of disability due to back pain as assessed using the Oswestry Disability Index (ODI) decreases in all study groups.
5. The functioning of patients in everyday life assessed using Back Pain Functional Scale (BPFS) shows the biggest improvement in the combination therapy group consisting of manual therapy and PNF method.

The research was conducted upon acquiring an approval of the Bioethical Commission of the Medical University in Lublin, no. KE-0254/14/2014.

The patients were qualified for physiotherapeutic treatment by a physician working in the given centre - a specialist on medical rehabilitation or balneotherapy and physical medicine. None of the physicians were personally involved in the preparation and the manner of performing the test. Before commencing the tests, each patient was familiarised with their course, expressed their informed written consent for participation and random assignment to a specified tested group.

The following selection criteria were used for the groups: the patients' age between 25 and 55, symptomatic spinal disc herniation of the lumbar spine at levels L4-L5 and/or L5-S1, confirmed by imaging examinations (MR, TK), and the patients' lack of use of non-steroid anti-inflammatory medications (NSAID) for a period of at least one week before the start of the tests. People with cancer, past spinal injuries with damage to anatomical structures and osteoporosis were excluded from the tests. Patients who did not qualify also included those with a severe inflammatory condition (sciatica) and a massive hernia of the intervertebral disc which required surgical treatment.

The patients were randomly divided into four groups of 50 people: the manual therapy group (A), the PNF method group (B), the manual therapy and the PNF method group (C), as well as the traditional kinesiotherapy group (D).

The analgesic effectiveness of the used treatment was evaluated by means of the commonly used 10-point visual analogue scale of pain (VAS), the modified Laitinen questionnaire of pain indicators and Stratford's back pain functional scale (BPFS).

The functional evaluation involved a goniometric examination of active movements in the thoracic-lumbar spine.

Measurements of pain intensity using visual analogue pain scale (VAS) and a modified Laitinen's pain questionnaire were carried out four times (before the start of the therapy, in the middle of it, immediately after its completion and two weeks after the end of the treatment). The clinical examination (goniometric measurements of range of motion of the spine) and the evaluation of nerve root irritation (Laseque's and Mackiewicz's signs) was done twice - before and immediately after the treatment. At the same time the functional ability in patients in everyday activities using the Oswestry Disability Index (ODI) and Back Pain Functional Scale (BPFS) was evaluated too.

A statistical analysis of the obtained results was performed by means of the SPSS 17.0 suite (Softonic, USA). Qualitative variables were described by stating the number of observations with the feature variant (n) and its corresponding percentage. The qualitative-nominal data was calculated using the following dependence coefficients based on the Chi-squared test: Phi, Cramér's V, and for variables in ordinal scales: Kendall's Tau-b and Tau-c.

Quantitative variables were described based on their nonparametric distribution (verified by the Shapiro-Wilk and Kolmogorov-Smirnov tests) by providing descriptive characteristics: the mean, the median and the standard deviation. In order to compare differences between analogical parameters, the following nonparametric tests were used: Mann-Whitney U for two samples, Kruskal-Wallis for more than two samples. In the case of comparisons of equinumerous groups, a parametric test was used: a one-way analysis of variance ANOVA and post hoc comparisons.

Inclusion Criteria:

• the patients' age between 25 and 55,
• symptomatic spinal disc herniation of the lumbar spine at levels L4-L5 and/or L5-S1,
• confirmed by imaging examinations (MR, TK),
• the patients' lack of use of non-steroid anti-inflammatory medications (NSAID) for a period of at least one week before the start of the tests.

Exclusion Criteria:

• people with cancer,
• past spinal injuries with damage to anatomical structures and osteoporosis,
• patients with a severe inflammatory condition (sciatica) and a massive hernia of the intervertebral disc which required surgical treatment.

None of the physicians were personally involved in the preparation and the manner of performing the test. Before commencing the tests, each patient was familiarised with their course, expressed their informed written consent for participation and random assignment to a specified tested group. The author of this paper had no influence on the categorisation of patients into individual treatment groups. During the studies, the physicians performing the categorisation were not informed about the results of the used treatment. However, on request they could acquire information about the result of using the procedures for a single patient. They were also informed about the occurrence of any undesired events.