This study evaluated the effect of smoking on the clinical response to non-surgical and surgical periodontal therapy. 74 adult subjects with moderate to advanced periodontitis were treated according to a split-mouth design involving the following treatment modalities: coronal scaling, root planing, modified Widman surgery, and flap with osseous resectional surgery. Clinical parameters assessed included probing depth, probing attachment level, horizontal attachment level in furcation sites, recession, presence of supragingival plaque and bleeding on probing. Data were collected: initially, 4 weeks following phase-I therapy, 10 weeks following phase-II therapy and on a yearly basis during 6 years of maintenance care. Data analysis demonstrated that smokers exhibited significantly less reduction of probing depth and less gain of probing attachment level when compared to non-smokers immediately following active therapy and during each of the 6 years of maintenance (p < 0.05). A greater loss of horizontal attachment level was evident in smokers at each yearly exam during maintenance therapy (p < 0.05). There were no differences between groups in recession changes. In general, these findings were true for the outcomes following all 4 modalities of therapy and were most pronounced in the deepest probing depth category (> or = 7 mm). Statistical analysis showed a tendency for smokers to have slightly more supragingival plaque and bleeding on probing. In summary, smokers responded less favorably than non-smokers to periodontal therapy which included 3-month maintenance follow-up.