A Petrobras deepwater exploration well is planned to be drilled in water depth greater than 2438 m (8,000 ft) offshore Brazil. As is typical of deepwater wells, it has a narrow drilling window between pore pressure and fracture gradient, requiring many casing strings to reach total depth (TD). Because pressure-related problems are often extremely costly when drilling conventionally in deep water, initial design assumptions were conservative, assuming the maximum pore pressure and minimum fracture gradient for casing-point and pipe selection. As with all conventional casing designs, the resulting pipe has limited capacity to withstand increased burst and collapse loads associated with extending hole sections if the opportunity arises.Managed-pressure drilling (MPD) enables the development of an adaptive well design by providing three key advantages over conventional drilling: the ability to detect kicks and losses extremely accurately (in gallons, rather than barrels); the ability to rapidly respond to and dynamically control detected kicks and losses by adjusting backpressure on the annulus, keeping kick volumes small; and the ability to perform dynamic pore-pressure and formation-integrity tests. Detecting and responding rapidly to kicks allows for the well to be designed with smaller kick tolerances. Furthermore, by keeping the kicks small, they can be circulated out quickly without requiring conventional, time-consuming well-control operations, which often lead to further hole problems. When this advantage is coupled with dynamic pore-pressure tests and dynamic formation-integrity tests, real-time optimization of casing points can be achieved safely with reduced risk of costly well-control incidents.The method presented here, applicable to any deepwater well, uses low, expected, and high pore-pressure and fracture-gradient estimates to develop an adaptive casing design for the well. In this paper, kick-tolerance calculations for selecting casing points by use of MPD are defined. The effect of reduced kick-tolerance requirements because of the application of MPD is demonstrated, and an adaptive procedure for real-time design optimization of the well is presented. The method results in decision criteria that identify key depths where casing strings can be eliminated without compromising the ability to reach objective TD. The resulting adaptive casing design will most likely eliminate at least one, and has the potential to eliminate two, casing strings, representing significant cost savings.